Process for controlling the energy performance of an industrial unit

• US 20060136100A1
• Filed: 12/13/2005
• Published: 06/22/2006
• Est. Priority Date: 12/17/2004
• Status: Active Grant

• Current Assignee: L'Air Liquide Societe Anonyme Pour L'Etude et L'Exploitation Des Procedes Georges Claude (L'Air Liquide SA)
• Original Assignee: I'Exploitation Des Procedes Georges Claude, L'Air Liquide Societe Anonyme Pour L'Etude et L'Exploitation Des Procedes Georges Claude (L'Air Liquide SA)
• Inventors: Drouart, Caroline, Roba, Thierry
• Granted Patent: US 8,073,654 B2
• Time in Patent Office: Days
• US Class Current: 700/273
• CPC Class Codes: G05B 17/02 electric
• CPC Class Codes: G05B 19/41885 characterised by modeling, simulation of the manufacturing system
• CPC Class Codes: G05B 2219/2639 Energy management, use maximum of cheap power, keep peak load low
• CPC Class Codes: G05B 2219/31414 Calculate amount of production energy, waste and toxic release
• CPC Class Codes: G06Q 10/06 Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
• CPC Class Codes: G06Q 50/04 Manufacturing
• CPC Class Codes: G06Q 50/06 Energy or water supply

Abstract

Process for controlling the energy performance of an industrial unit, in which process: the value of a set (V) of variables relating to the industrial unit and to the environment are measured, as is the value of a quantity representing the energy performance (W), an estimate (W_e) of the quantity representing the energy performance is calculated, the value of an energy performance indicator (TE, TE₊) is determined and the value of this indicator is compared with at least one threshold value (S) determined so as to detect a drift in the energy performance. The invention proposes the learning of a model based on the best energy performance (W) observed for each value of a set (V) of input variables and in the proposing in real time to the operator of a best configuration and of best settings (CR) of the equipment based on those associated with this best energy performance (W) observed in the past for the current value of the set (V) of input variables.

Claims

1. A process for controlling the energy performance of an industrial unit or of a part of the latter, in which process:

a) the value of a set (V) of variables relating to the production of the industrial unit and to the environment are measured, as is the value of a quantity representing the energy performance (W);

b) an estimate (W_e) of the quantity representing the energy performance is calculated, using a statistical model (ML, MNL) calibrated during a phase of learning over a reference period, on the basis of the values of the set (V) of the variables relating to the production of the industrial unit (ASU) and to the environment;

c) the value of an energy performance indicator (TE, TE₊) is determined on the basis of the measured value (W) and estimated value (W_e) of the quantity representing the energy performance; and

d) the value of this indicator is compared with at least one threshold value (S) determined so as to detect a drift in the energy performance.

2. The process of claim 1, wherein the quantity representative of the energy performance is the electrical power consumed (W) of the industrial unit.

3. The process of claim 1, wherein an alert to an operator is emitted in case of energy over-consumption.

4. The process of claim 1, wherein said energy performance indicator (TE, TE₊) is defined as the ratio of the measured value (W) to the estimated value (W_e) of the quantity representing the energy performance.

5. The process of claim 1, wherein the statistical model is of linear type (ML), in particular obtained by linear regression on data of the learning phase.

6. The process of claim 1, wherein the statistical model is of nonlinear type (MLN), in particular comprising at least one neural network or tools for local regression.

7. The process of claim 1, wherein, during the learning phase, the best operating points, corresponding to the best energy performance (W) observed associated with each given value of the set (V) of input variables, are taken into account in the model (MNL).

8. The process of claim 7, wherein in the model (MNL), each association (ASSO) between a given value of the set (V) of input variables and the best energy performance (W), corresponding to an operating point, also comprises the set (CR) of configuration and setting data for the equipment associated with this operating point.

9. The process of claim 7, wherein, when an over-consumption is detected, the model (MNL) provides a set (CR+) of learnt configuration and setting data for the equipment that made it possible to obtain the best energy performance (W_e⁺) for the current value of the set (V) of input variables.

10. The process of claim 1, wherein the set (V) of input variables comprises:

a) the atmospheric pressure (P_a) and the ambient temperature (T_a);

b) production flow rates (QN_g^np, QO_g^hp, QO_g^mp, QAir, QN_l, QO_l) for the unit;

c) production pressures for the plant (PN_g^hp, PO_g^hp, PO_g^mp, PAir);

d) production rating data (PPO_g^mp); and

e) flow rates of vaporization (QN_v,QN_o) of the liquids and of venting of gases,

said set (CR) of configuration and setting data for the equipment comprises the configuration of the equipment and their setting in terms of flow rate, and the energy performance (W) is that of the industrial unit as a whole.

11. The process of claim 1, wherein said set (V) of input variables comprises:

a) the flow rate and the pressure of the gas produced from a group of equipment of compressor type; and

b) the pressure and the temperature upstream of the compressors,

said set (CR) of configuration and setting data for the equipment comprises the configuration of the equipment and their setting in terms of flow rate, and the energy performance (W) is that of the group of compressors.

12. The process of claim 1, wherein said set (V) of input variables comprises:

a) the liquid flow rate of product (N_L) leaving a liquefier; and

b) the pressure and the temperature of the gas entering the liquefier,

said set (CR) of configuration and setting data for the equipment comprises the pressure of the cycle, the temperature of the liquid and the recycling flow rate, and the energy performance (W) is that of the liquefier.

13. The process of claim 8, wherein said set (CR) of configuration and setting data for the equipment of the industrial unit comprises at least one variable of usage of an item of equipment indicating its on or off state (UC_F, UP, UO).

14. The process of claim 8, wherein said set (CR) of configuration and setting data for the equipment of the industrial unit comprises at least one value of setting of an item of equipment, in particular a flow rate setting (QC_v, QP, QO).

15. The process of claim 13, wherein the set (CR+) of learnt configuration and setting data for the equipment that made it possible to obtain the best energy performance (W_e+) for the current value of said set (V) of input variables is proposed to the user, with:

a) the list of advocated equipment, using the on/off values of the usage variables (UC_v, UC_F, UP, UO) recorded; and

b) the estimate of the settings (QC_v, QP, QO) for the equipment, if necessary.

16. The process of claim 15, wherein a list of the changes induced so as to achieve a given configuration and settings of the equipment (CR₊) with respect to the current configuration and settings (CR) is provided, indicating the equipment to be turned off or to be started up, as well as the possible settings to be effected.

17. The process of claim 1, wherein the process is implemented in real time, the measurements being performed continuously.

18. The process of claim 1, wherein a recalibration is performed, so as to integrate new learning data, at a pre-established frequency, or on the request of an operator.

19. A system for controlling the energy performance of an industrial unit, making it possible to implement a process of claim 1, wherein it comprises:

a) means of measurement in the industrial unit;

b) means of storage making it possible to record the corresponding data;

c) means of calculation;

d) means of informing an operator; and

e) means of network communication linking the components of the control system.