Method of controlling supply of power

• US 3,932,735 A
• Filed: 02/03/1972
• Issued: 01/13/1976
• Est. Priority Date: 08/24/1970
• Status: Expired due to Term

• Current Assignee: Westinghouse Electric Company LLC (Brookfield Asset Management Incorporated)
• Original Assignee: Westinghouse Electric Company LLC (Brookfield Asset Management Incorporated)
• Inventors: Giras, Theodore C.
• Primary Examiner(s): Morrison, Malcolm A.
• Assistant Examiner(s): Wise, Edward J.
• Application Number:
  US05/223,378
• Time in Patent Office: 1,440 Days
• Field of Search: 235/151.21, 444/1
• US Class Current: 700/287
• CPC Class Codes: H02J 3/00 Circuit arrangements for ac mains or ac distribution networks

Abstract

A digital computer system is employed to produce power system control. The computer system employs an economic dispatch program, a unit selection program and a tie line interchange program to determine off line a tie line schedule and unit commitments needed to satisfy economically and safely a load forecast over a selected time interval such as 24 hours. In on line closed control loop operation, the computer system employs a load and frequency control program which uses the tie line schedule and other quantities in determining area control error. The control program develops a system error representation which reflects a smoothed present and anticipated area control error. The system error representation is distributed into individual generator plant control signals under economic dispatch program and unit selection program control and under constraint of programmed plant limits to assure that each generation move is error correcting as well as optimally economic.

Claims

1. The method of, operating a power complex to supply a system load and non-conforming loads;

said power complex having a power generating network including a plurality of power generators and tie-lines to neighboring power generating networks, the said method comprising deriving from said complex at relatively short intervals, of the order of one second, signals indicative of;

1. the load power supplied by each of said generators,

2. the frequency of the power of said load,

3. the power drawn by said tie-lines,

4. the power drawn by said non-conforming loads;

the said method also comprising converting said signals into digital values, separating said digital values into separate items of data each item corresponding to said items 1, 2, 3 and 4 respectively;

storing said separate items;

computing from a plurality of said items derived during a plurality of said intervals and the past history of operation of said complex, an area-requirement error which is based, for each computator on the anticipated trend of variation of said system load following the latest of said intervals, responsive to said area-requirement error producing electrical generator resetting signals to reset economically each power generator of said plurality of generators to the extent required to reduce said error to zero, and supplying said resetting signals to said generators to reset said generators as aforesaid.

2. A system for controlling a power system having a plurality of power generating units, said system comprising means for determining a representation of an actual power system generation error, means for determining a smoothed anticipatory system power generation error representation as a function of the actual system error representation, and means for controlling selected generating units in the power system as a function of the smoothed anticipatory system error representation.

3. A system for generating power in accordance with schedules of system frequency and of power interchange with a neighboring power system said system comprising:

a. a plurality of generating units operative selectively to generate power in accordance with the schedule of power interchange, at least one tie-line to the neighboring power system, and means for measuring and providing representations of the tie-line power, the frequency deviation of the power generated by the given system with respect to the scheduled system frequency, and the power generated by each of the plurality of units;

b. instantaneous error detecting means responsive to the representations of tie-line power, the schedule of power interchange and of frequency deviation for providing a representation of instantaneous system area-control error;

c. calculation means for smoothing and weighting the representation of instantaneous system area-control error to provide an anticipated system power generating error; and

d. unit control means for applying control signals to each of said generating units as a function of the anticipated system power generating error whereby the schedule of power generation is met.

4. A system as claimed in claim 3, wherein said system further includes sampling means for sampling at intervals of time n, (n - T_S) and (n = 2T_S), where T_S is a sampling time interval, the tie-line power, the frequency of the given system and the power generated by each of said plurality of generating units.

5. A system as claimed in claim 4, wherein said calculation means provides a representation of anticipated system power generating error for the interval (n + T_P), where T_P is an anticipatory interval of time.

6. The system as claimed in claim 5, wherein said interval T_P is greater than said interval T_S.

7. The system as claimed in claim 6, wherein said unit control means includes means for measuring the power demanded of each of said plurality of power generating units dependent upon a selected percentage of the anticipated system power generating error for determining the incremental change of power demanded of that generating unit in the interval (n + T_P).

8. The system as claimed in claim 3, wherein said instantaneous error detecting means provides the representation of instantaneous system area-control error as a function of the sum of the representation of the tie-line power flow and the system frequency deviation modified by a value B indicative of the power variation of the given system per unit deviation of frequency, less the scheduled power demanded of the system.

9. The system as claimed in claim 8, wherein said calculation means includes a first intermediate calculating means for calculating a first intermediate error as a function of the sum of the instantaneous system area-control error as derived from said instantaneous error detecting means and the power required of all of said power generating units, less the measured power generated by all of said generating units.

10. The system as claimed in claim 9, wherein said calculation means includes a second intermediate calculating means for providing a second intermediate error as a function of the first intermediate error modified by a factor K_S indicative of the power generating gain of the given power system to tend thereby to smooth the effect of the instantaneous system area-control error upon the determination of the anticipated system power generating error.

11. The system as claimed in claim 10, wherein said calculation means includes means for calculating the weighted average of the system power generating error for the interval n as a function of the sum of the weighted average system error for the preceding interval (n - T_S) and the product of the rate of change of the system power generating error for the previous interval (n - T_S) and the interval T_S, plus the second intermediate error, the aforementioned sum and the intermediate error being modified by selected weighting coefficients calculated to damp critically the power generation of the given system and to minimize noise.

12. The system as claimed in claim 11, wherein said calculation means further includes means for deriving the time rate of change of the system power generating error for the interval n as a function of the time rate of change of the system power generating error for the previous interval (n - T_S), and the second intermediate error less the weighted average system power generating error for the previous interval (n - T_S), the aforementioned values being modified by weighting coefficients calculated to critically damp the power generation of said plurality of units and to minimize noise.

13. The system as claimed in claim 12, wherein said calculation means provides a representation of the anticipated system power generating error for the interval (n + T_P) as the sum of the weighted average system power generating error for the interval n and the product of the time rate of change of the system power generating error for the interval n and the interval T_P.

14. The system as claimed in claim 13, wherein said unit control means includes means for providing a representation of the power demanded of each of said plurality of generating units dependent upon the percentage of participation by one of said generating units to meet the anticipated system power generating error for the interval (n + T_P), and upon the base power output of that turbine for the interval n.

15. The system as claimed in claim 14, wherein there is further included storage means for storing and providing representations of the incremental costs of power generation of each of said plurality of generating units, and comparison means for comparing the incremental cost of power generation as provided by said storage means for causing said unit control means to establish an economic condition wherein the base output of each of said plurality of generating units is set so that the incremental costs of power generation of each of said plurality of generating units tends to be equal.

16. The system as claimed in claim 15, wherein said unit control means includes override means responsive to the condition where the total power demanded of all of said generating units exceeds the power generated by said plurality of generating units at said economic condition, to cause said unit control means to increase the power generation of selected generating units.

17. The system as claimed in claim 14, wherein said unit control means further includes means for calculating the instantaneous power generating error for one of said generating units for the interval n as a function of the difference of the power demanded of said one generating unit and the actual power generated by said one generating unit modified by a factor K indicative of the power generating gain of that generating unit.

18. The system as claimed in claim 17, wherein said unit control means includes means for providing a representation of the weighted average of the power generating error of said one generating unit for the interval n as a function of the sum of the weighted average power generating error for said one generating unit for the previous interval (n - T_S), and the product of T_S and the time rate of change of the power generating error for said one generating unit for the previous interval (n - T_S), and as a function of the instantaneous power generating error for said one generating unit for the interval n, said aforementioned values being modified by weighting coefficients selected to critically damp the power generation of the given system and to minimize noise.

19. The system as claimed in claim 18, wherein said unit control means includes means for providing a representation of the time rate of change of the power generating error for said one generating unit for the interval n as a function of the time rate of change of the power generating error for said one generating unit for the previous interval (n - T_S) and as a function of the difference between the instantaneous power generating error for said one unit for the interval n and the weighted average of the power generating error for said one generating unit for the previous interval (n - T_S), said aforementioned representation being modified by coefficients selected to damp critically the power generation of the given system and to minimize noise.

20. The system as claimed in claim 19, wherein said unit control means provides a representation of the anticipated power generating error for said one generating unit as a function of the sum of the weighted average power generating error for said one generating unit for the interval n and the product of the time rate of change of the power generating error for said one generating unit for the interval n and the interval T_P.

21. The system as claimed in claim 20, wherein there is further included storage means for storing and providing the weighted average system power generating error for the preceding interval (n - T_S), the time rate of change of the system power generating error for the preceding interval (n - T_S) the weighted average of the power generating error for each of said generating units for the preceding interval (n - T_S) and the time rate of change of the power generating error for each of said generating units for the preceding interval (n - T_S).

22. The system as claimed in claim 21 wherein said storage means further stores and provides representations of the factor K_S, the weighting coefficients, and the factors K for each of said generating units.

23. The system as claimed in claim 19, wherein there is included storage means for storing and providing representations of the rate limits of power generation for each of said plurality of generating units, and comparison means for comparing the same rate of change of the power generating error for one of said generating units for the interval n with the rate limit for said one generating unit as provided by said storage means, for preventing said control unit means from effecting a greater rate of power generation change from said one generating unit than its rate limit.

24. The system as claimed in claim 12, wherein there is further included storage means for storing and providing representations of the weighted average system power generating error for the preceding interval (n - T_S) and the time rate of change of the system power generating error for the previous interval (n - T_S).

25. A system for controlling the power generation of a given power system in accordance with a scheduled frequency, the given system including a plurality of generating units, and means for monitoring the frequency deviation of the power generated by the given system with respect to the scheduled frequency and the power generated by each of the plurality of generating units, said control system comprising:

a. instantaneous error detecting means responsive to the representation of frequency deviation from the scheduled frequency for providing a representation of instantaneous system area-control error;

b. calculation means for smoothing and weighting the representation of instantaneous system area-control error to provide an anticipated system power generating error; and

c. unit control means for controlling each of the power generating units as a function of the anticipated system power generating error.

26. A control process for controlling a given power system having a plurality of power generating units and at least one tie-line therefrom to a neighboring power system, said control process comprising the steps of:

a. measuring representations of the tie-line power, the frequency deviation of the power generated by the given power system with respect to a scheduled frequency and the power generated by each of the plurality of generating units;

b. calculating the instantaneous system area-control error as a function of the tie-line power, a schedule of power interchange with the neighboring systems and the frequency deviation;

c. smoothing and weighting a representation of instantaneous system area-control error to provide an anticipated system power generating error, said steps of calculating, and smoothing and weighting being carried out in a digital computer; and

d. applying control signals to each of the generating units as a function of the anticipated system power generating error.

27. The control process as claimed in claim 26, wherein the step of determining in accordance with paragraph (a) includes the sampling for intervals of time n, (n -T_S), (n - 2T_S) of the tie-line power, the frequency deviation of the given system and the power generated by each of the generating units, where T_S is a sampling interval.

28. The control process as claimed in claim 26, wherein the step in accordance with paragraph (b) includes the calculation of the instantaneous system area-control error E_A (n) in accordance with the following equation:

space="preserve" listing-type="equation">E.sub.A (n) = Σ

P.sub.Ti (n) - P.sub.S (n) + BΔ

F(n)

where B is the power variation per unit deviation in frequency of the given system.

29. The contnrol process as claimed in claim 28, further including the step of:

determining the power P_NCi (n) required by non-conforming loads associated with the given system, and

said step of calculating the instantaneous system area-control error E_A (n) further includes a factor indicative of the anticipated sum of the required power of all the non-conforming loads Σ

P_NCi (n).

30. The control process as claimed in claim 28, wherein the step of smoothing and weighting in accordance with paragraph (c) includes the steps of:

summing the power demanded of all of the generating units for the computer solution interval n to provide a value Σ

P_D (n);

summing the power generated by all of the generating units for the computer solution interval n to provide a value Σ

P_A (n); and

calculating a first intermediate error E_I (n) in accordance with the following formula;

space="preserve" listing-type="equation">E.sub.I (n) = E.sub.A (n) - Σ

P.sub.A (n) + Σ

P.sub.D (n)

31. The control process as claimed in claim 30, wherein the step of smoothing and weighting in accordance with paragraph (c) further includes:

the storing and providing of the value E_O (n - T_S) indicative of a second intermediate error for the previous interval (n - T_S), and

the calculation of a second intermediate error E_O (n) for the computer solution interval n in accordance with the equation;

space="preserve" listing-type="equation">E.sub.O (n) = K.sub.S E.sub.I (n) + E.sub.O (n - T.sub.S)

where K_S is indicative of the overall power gain of the given system.

32. The control process as claimed in claim 31, wherein the step of smoothing and weighting in accordance with paragraph (c) further includes:

the storing and providing the representations of the weighted system power generating error E_S (n - T_S) for the previous interval (n - T_S), and the time rate of change of the system power generating error E_S (n - T_S) for the previous interval (n - T_S); and

the calculation of the weighted average system power generating error E_S (n) for the computer solution interval n in accordance with the following formula;

space="preserve" listing-type="equation">E.sub.S (n) = (1 - α

)[E.sub.S (n - T.sub.S) + T.sub.S E.sub.S (n - T.sub.S)]+ α

E.sub.O (n)

where α

is a weighting coefficient.

33. The control process as claimed in claim 32, wherein the step of smoothing and weighting in accordance with paragraph (c) further includes:

the calculation of the time rate of change of the system power generating error E_S (n) for the computer solution interval n in accordance with the following equation;

##EQU3## where β

is a weighting coefficient.

34. The control process as claimed in claim 33, wherein the step of smoothing and weighting in accordance with paragraph (c) further includes:

the calculation of the values of α and

β

to critically damp the power generation of the given system in accordance with the following equation

space="preserve" listing-type="equation">4β

- α

.sup.2 - β

.sup.2 - 2α

β

= 0.

35. The control process as claimed in claim 34, wherein the step of smoothing and weighting in accordance with paragraph (c) further includes:

the calculation of the values of α and

β

to provide minimum system noise in accordance with the following equation;

##EQU4##

36. The control process as claimed in claim 35, wherein the step of smoothing and weighting in accordance with paragraph (c) further includes:

the calculation of the anticipated system power generating error E_S (n + T_P) for the interval (n + T_P) in accordance with the following equation;

space="preserve" listing-type="equation">E.sub.S (n + T.sub.P) = E.sub.S (n) + T.sub.P E.sub.S (n)

37. The control process as claimed in claim 26, wherein the step of controlling each generating unit in accordance with paragraph (d) further includes:

providing a representation of the base powers B_i (n) for all of the generating units for the computer solution interval n;

providing a representation of the percentages (%) participation of all of the generating units to meet the demanded incremental change of power in the interval (n + T_P); and

calculating the power P_Di (n) required of each of the plurality of generators for the computer solution interval n in accordance with the following equation;

space="preserve" listing-type="equation">P.sub.Di (n) = E.sub.S (n + T.sub.P)(%).sub.i + Base.sub.i (n)

38. The control process as claimed in claim 37, wherein the step of controlling each generating unit in accordance with paragraph (d) further includes:

the computation of the Base_i (n) for each of the plurality of generating units in accordance with an economic dispatch program to achieve an economic condition wherein the incremental power generating cost of each of the generating units is approximately equal.

39. The control process as claimed in claim 37, wherein the step of controlling each generating unit in accordance with paragraph (d) further includes:

providing a representation of power generated P_A (n) for each of the generating units for the computer solution interval n; and

the calculation of the instantaneous error E(n) of each generating unit for the computer solution interval n in accordance with the following equation;

space="preserve" listing-type="equation">E(n) = K[P.sub.D (n) - P.sub.A (n)]

where K is indicative of the gain of the generating unit regulating loop.

40. The control process as claimed in claim 39, wherein the step of controlling the generating units in accordance with the paragraph (d) further includes:

the storing and providing of representations indicative of the weighted average power generating error E_Q (n - T_S) for each generating unit for the previous interval (n - T_S) and the time rate of change of the power generating error E_Q (n - T_S) for the previous interval (n - T_S); and

the calculation of the weighted average power generating error E_Q (n) of each generating unit for the computer solution interval n in accordance with the following equation;

space="preserve" listing-type="equation">E.sub.Q (n) = (1 - α

)[E.sub.Q (n - T.sub.S) + T.sub.S E.sub.Q (n - T.sub.S)] + α

E(n)

where α

is a weighting coefficient.

41. The control process as claimed in claim 40, wherein the step of controlling the generating units in accordance with the paragraph (d) further includes:

the calculation of the time rate of change of the power generating error E_Q (n) of each generating unit for the computer solution interval n in accordance with the following equation;

##EQU5## where β

is a weighting coefficient.

42. The control process as claimed in claim 41, wherein the weighting coefficients α

and β

are calculated to solve the following equations;

##EQU6## to thereby critically damp the generation of power by the given system and to minimize system noise.

43. The control process as claimed in claim 41, wherein the step of controlling the generating units in accordance with paragraph (d) is effected to compensate for anticipated power generating error E_Q (n + T_P) for the interval (n + T_P) in accordance with the following equation:

space="preserve" listing-type="equation">E.sub.Q (n + T.sub.P) = E.sub.Q (n) + T.sub.P E.sub.Q (n)

where T_P is an anticipatory interval.

44. The control process as claimed in claim 43, wherein the anticipatory interval T_P is greater than the sampling interval T_S.

45. The control process as claimed in claim 41, wherein the step of controlling the generating units in accordance with paragraph (d) further includes:

providing a representation of the rate limit of change of power generation change for each generating unit; and

comparing the time rate of change of the power generating error E_Q (n) for each generating unit with the predetermined rate limit to prevent the change of the time rate of the power generation of a unit generator from exceeding its rate limit.

46. The control process as claimed in claim 26 wherein the step of controlling the power generation of each generating unit in accordance with paragraph (d) includes the step of selecting a combination of the plurality of generating units to achieve the most economic power generation to meet the anticipated system power generating error in accordance with a unit commitment program, said unit commitment program including the steps of:

determining the possible combination of generating units for the nth interval T_n, where T is a predetermined interval;

determining the production cost for the interval T_n for every combination of generating units in the interval T_n ;

determining the combination of generating units employed in the previous interval T_n-1, and calculating the start-up cost of the additional generating units required for the combination in the present interval T_n ;

determining the combination of generating units for the interval T_n-1 which minimizes the system cost of generating power from the first interval T₁ to the combination of power generating units in the present interval T_n and storing this determined combination and its power generating cost; and

determining the most economical combination of generating units in each interval T and the power generating cost for each interval T.

47. The control process as claimed in claim 46, wherein an economic dispatch program includes the step of determining the most economical settings of Base_i (n) and percentage (%) for each generating unit.

48. A system for controlling the power generation of a given power system in accordance with a schedule of power interchange with a neighboring power system which is P_S (n) for the interval n, the given system including a plurality of generating units, at least one tie-line to the neighboring power system, and means for measuring and providing representations of the tie-line power P_Ti (n) for the interval n, the frequency deviation Δ

F of the power generated by the given system with respect to a scheduled frequency, and the power P_Ai (n) generated by each of the plurality of units for the interval n, said system comprising;

a. instantaneous error detecting means responsive to the representations of tie-line power P_Ti (n), of the scheduled power interchange P_S (n) and of frequency deviation Δ

F for providing a representation of the instantaneous system area-control error E_A (n) in accordance with the following equation;

space="preserve" listing-type="equation">E.sub.A (n) = Σ

P.sub.Ti (n) - P.sub.S (n) + BΔ

F(n)

where B is a factor indicative of the system power variation per unit deviation in frequency;

b. calculation means for smoothing and weighting the representation of instantaneous system area-control error to provide an anticipated system power generating error E_S (n + T_P) for the interval (n + T_P), said error detecting and calculation means including a digital computer; and

c. unit control means for applying control signals to each of the generating units as a function of the anticipated system power generating error E_S (n + T_P).

49. The system as claimed in claim 48, wherein there is further included scanning means associated with the measuring means, for scanning sequentially the representations of tie-line power P_Ti (n) and the frequency deviation of the given system whereby the instantaneous values of these representations are averaged over a sampling interval T_S.

50. The system as claimed in claim 48, wherein there is further included scanning means associated with the measuring means, for multiplexing the representations of tie-line power and frequency deviation whereby the effect of short-term variations of these representations upon the power generation of the given system is minimized.

51. The system as claimed in claim 48, wherein said calculation means further includes:

means for providing a representation of the power P_Di (n) demanded for each generating unit for the interval n and for providing a representation Σ

P_D (n) indicative of the sum of the demanded power of all of the generating units for the interval n;

means for providing a representation Σ

P_A (n) indicative of the sum of the power generated by all of the generating units for the interval n; and

means for calculating a first intermediate error E_I (n) in accordance with the following equation;

space="preserve" listing-type="equation">E.sub.I (n) = E.sub.A (n) - Σ

P.sub.A (n) +Σ

P.sub.D (n)

52. The system as claimed in claim 51, wherein said calculation means further includes:

means for storing and providing a representation of a second intermediate error E_O (n - T_S) for the previous interval (n - T_S), wherein T_S is a sampling interval; and

means for providing a representation of the second intermediate error E_O (n) for the interval n in accordance with the following equation;

space="preserve" listing-type="equation">E.sub.O (n) = K.sub.S E.sub.I (n) + E.sub.O (n - T.sub.S)

where K_S is a power generating gain factor of the given system.

53. The system as claimed in claim 52, wherein said calculation means further includes:

storage means for storing and providing a representation of the weighted average system power generating error E_S (n - T_S) for the previous interval (n - T_S) and the time rate of change of the system power generating error E_S (n - T_S) for the previous interval (n - T_S); and

means for calculating the weighted average system power generating error E_S (n) for the interval n in accordance with the following equation;

space="preserve" listing-type="equation">E.sub.S (n) = (1 -α

) [E.sub.S (n - T.sub.S) + T.sub.S E.sub.S (n - T.sub.S)] + α

E.sub.O (n)

where α

is a weighting coefficient.

54. The system as claimed in claim 53, wherein said calculation means further includes:

means for calculating the time rate of change of the system power generating error E_S (n) for the interval n in accordance with the following equation;

##EQU7## where β

is a weighting coefficient.

55. The system as claimed in claim 54, wherein said calculation means further includes:

means for calculating the anticipated system power generating error E_S (n + T_P) in accordance with the following equation;

space="preserve" listing-type="equation">E.sub.S (n + T.sub.P) = E.sub.S (n) + T.sub.P E.sub.S (n)

where T_P is an anticipatory interval.

56. The system as claimed in claim 55, wherein said calculation means further includes:

means for calculating the values of α and

β

to solve the following equations;

##EQU8## to critically damper the power generation of this system and to minimize system noise.

57. The system as claimed in claim 55, wherein said unit control means further includes:

means for storing and providing representations of the desired power generating output Base_i for each generating unit and of the percentage (%) of participation of each of the generating units in a demanded incremental change of system power; and

means for calculating the power demanded of each generating unit for the interval n in accordance with the following equation;

space="preserve" listing-type="equation">P.sub.Di (n) = E.sub.S (n + T.sub.P) (%).sub.i + Base .sub.i (n)

58. The system as claimed in claim 57, wherein said unit control means further includes:

means for calculating in accordance with an economic dispatch program values of Base_i (n) for each of the generating units such that the incremental cost of power generation of each generating unit is approximately equal.

59. The system as claimed in claim 57, wherein the plurality of generating units is disposed within a network interconnecting the plurality of generating units to a load, said unit control means further including:

means for calculating values of base power output Base_i (n) for each generating unit in accordance with a set of penalty factors indicative of the power loss due to its transmission from one of the generating units through a portion of the interconnecting network to the load.

60. The system as claimed in claim 57, wherein said unit control means further includes:

means for calculating an instantaneous error E(n) for each generating unit for the interval n in accordance with the following equation;

space="preserve" listing-type="equation">E(n) = K[P.sub.D (n) - P.sub.A (n)]

where K is a factor indicative of the power generation gain of that generating unit.

61. The system as claimed in claim 60, wherein said unit control means further includes:

storage means for storing and providing representations of the weighted average power generating error E_Q (n - T_S) of each generating unit for the previous interval (n - T_S), and the time rate of change of the power generating error E_Q (n - T_S) of each generating unit for the previous interval (n - T_S); and

means for calculating the weighted average power generating error E_Q (n) of each generating unit for the interval n in accordance with the following equation;

space="preserve" listing-type="equation">E.sub.Q (n) = (1 - α

) [E.sub.Q (n - T.sub.S) + T.sub.S E.sub.Q (n - T.sub.S)] + α

E(n)

where α

is a weighting coefficient.

62. The system as claimed in claim 61, wherein said unit control means further includes:

means for calculating the time rate of change of the power generating error E_Q (n) of each generating unit for the interval n in accordance with the following equation;

##EQU9## where β

is a weighting coefficient.

63. The system as claimed in claim 62, wherein said unit control means further includes:

means for calculating the anticipated system power generating error E_Q (n + T_P) of each generating unit for the interval (n + T_P) in accordance with the following equation;

space="preserve" listing-type="equation">E.sub.Q (n + T.sub.P) = E.sub.Q (n) + T.sub.P E.sub.Q (n)

where T_P is an anticipatory interval.

64. The system as claimed in claim 63, wherein there is included alarm means responsive to the anticipated system power generating error E_S (n + T_P) and the anticipated power generating error E_Q (n + T_P) of each generating unit to provide an alarm signal indicative of the condition where the power generating reserve of the given system is inadequate to meet the anticipated power demands on the given system and/or on one or more of the generating units.

65. The system as claimed in claim 63, wherein the anticipatory interval T_P is greater than the sampling interval T_S.

66. The system as claimed in claim 63, wherein said unit control means includes:

means for storing and providing representations of the permissible loading and unloading rates of each of the generating units; and

comparison means for comparing for each generating unit its anticipated power generating error E_Q (n + T_P) and the stored values of its loading and unloading rates to prevent said unit control means from loading or unloading a generating unit at a rate greater than its permissible loading and unloading rates.

67. The system as claimed in claim 62, wherein said unit control means further includes:

storage means for storing and providing representations of the rate limit of change of power generation of each of the generating units;

comparison means for comparing the calculated time rate of change of power generation E_Q (n) of each generating unit for the interval n with the stored rate limit for each generating unit, to prevent said unit control means from causing a rate change of power of a generating unit greater than its rate limit.