BR102015017609A2 - system and method for a load anticipation feature and its throttling method for a generation set - Google Patents

Abstract

Summary “system and method for a load anticipation feature and its tuning method for a generation set” is a computer method, system, and software for providing a load anticipation feature for a diesel generator set which includes a diesel engine, a generator, a speed governor and an automatic voltage regulator generator in a power system, a computer data processing apparatus and computer executable instructions for determining an equivalent linear speed control system whose response corresponds approximately to the nonlinear speed response of the diesel generator set; determining at least one programmable load anticipation feature parameter based on an actual power load applied to the diesel generator set and a desired speed response from the linear speed control system; generate a control output based on a measured actual power load and at least one programmable load anticipation parameter; wherein the speed governor receives the control output and adjusts the speed of the diesel engine based on the control output. 1/1

Description

“SYSTEM AND METHOD FOR A LOAD ADVANCE FEATURE AND ITS REGULATION METHOD FOR A GENERATION SET” CROSS REFERENCE TO RELATED APPLICATIONS

This application claims priority to Provisional Patent Application No. 62 / 028,653, filed July 24, 2014 and Provisional Patent Application No. 62 / 03,482 filed August 5, 2014, wherein the disclosure of each one is incorporated herein by reference in its entirety.

FIELD OF TECHNIQUE

The present disclosure relates to alternating current power generators and more specifically to a system and method for load application performance of a diesel generator set.

BACKGROUND

Gensets are used to supply electricity to distributed power generation systems, which include primary power, standby generation and grid support. Deregulation of power utilities has resulted in many consumers using their standby diesel generator sets to improve power quality or to avoid peak electricity rates.

Generator sets generally consist of a diesel engine, a synchronous machine, and two controllers (one speed governor and one automatic voltage regulator (“AVR”)). Some of the modern microprocessor-based AVRs are deployed with Proportional, Integral, and Derivative (PID) control for stabilization and various supplemental control systems that include many limiters, VAR / PF controller, throttling functionality, protection, and monitoring features.

[0005] The generator set output frequency is directly proportional to the diesel engine speed. The governor tries to keep the diesel engine at a constant speed. The governor is designed to respond to changes in engine speed by acting as a feedback controller that controls the fuel rate to minimize a speed deviation due to a sudden change in the actual generator set power load. Unlike large size generators, many generator sets can be expected to change from unloading to fully loaded operation in a single-step load application, which causes large changes in engine speed or may cause the engine to late.

[0006] The International Standard ISO8528-5; 2013 provides a means to evaluate diesel generator set performance, which classifies a generator set based on a series of key performance indicators. For modern G2 rated generator sets, the maximum voltage deviation from the nominal setpoint for sudden load acceptance shall not exceed 20% and the maximum electrical frequency deviation shall not exceed 10%. Voltage recovery time should be less than 6 seconds and frequency recovery time should be less than 5 seconds. Although some diesel generator set controllers are capable of meeting the requirements of ISO8528-5; 2013, such controllers are expensive to manufacture.

[0007] A common way to reduce engine speed drop is to provide an additional voltage DIP during speed drops. In this way, faster engine speed recovery is achieved by reducing the actual power load. Several voltage setpoint adjustments (schemes below frequency) are used in modern AVRs. A load adjustment scheme has been suggested that temporarily reduces the voltage during a transient and thus assists in motor speed recovery. However, the speed governor reaction to a motor speed change is much slower than the actual power change. As such, the inventor has identified a need for a system and method that can reduce the engine speed deviation of a generator set when the actual power load changes.

SUMMARY OF THE INVENTION

The present invention provides a computer method, system and software for providing a load anticipation feature for a diesel generator set that includes a diesel engine, a generator, a speed governor and an automatic voltage regulator. which includes the functions of determining an equivalent linear speed control system, the linear speed control system response that roughly corresponds to the nonlinear speed response of the diesel generator set, and determining at least one programmable parameter of a load anticipation based on a real power load applied to the diesel generator set and a desired speed response from the linear speed control system.

The system and method may also include the functions of measuring a real power load applied to the diesel generator set by a sensor electrically coupled to the diesel generator set, providing at least one programmable parameter of the anticipation feature. load for a computer data processing apparatus having computer memory, wherein the computer memory contains computer executable instructions for deploying the equivalent linear speed control system and the load anticipation feature, wherein the executable instructions Computer-assisted load anticipation features include instructions for generating a control output based on the actual measured power load and at least one programmable load anticipation parameter. The system and method may further include providing the control output to the governor, wherein the governor adjusts the speed of the diesel engine based on the control output.

BRIEF DESCRIPTION OF DRAWINGS

[0010] Figure 1 is a block diagram of a diesel generator set according to an exemplary embodiment.

[0011] Figure 2 is a functional block diagram showing the interaction between voltage control and speed control of the diesel generator set according to an exemplary embodiment.

[0012] Figure 3 is a functional block diagram of a model of an exemplary embodiment of an equivalent linear speed control system.

[0013] Figure 4 is a block diagram of a procedure that can be used to identify the programmable parameters of the equivalent linear speed control system according to an exemplary embodiment.

[0014] Figure 5 is a flow chart of a system and method used to determine an equivalent linear control system for the diesel generator set using Particle Swarm Enhancement, according to an exemplary embodiment.

[0015] Figure 6 is a functional block diagram of a Load Application Feature (LAF) modifier for an equivalent linear speed control system according to an exemplary embodiment.

[0016] Figure 7 is a block diagram illustrating a procedure for estimating the programmable LAF parameters according to an exemplary embodiment.

[0017] Figure 8 is a flowchart of a system and method used to estimate programmable parameters for LAF using Particle Swarm Enhancement according to an exemplary embodiment.

Figures 9A and 9B are functional block diagrams of an alternator-rectifier motor and driver representation according to an exemplary embodiment. Figure 10 shows an actual recorded speed response for real power. [0020] Figure 11 shows the response time of an equivalent linear speed control system compared to the response time predicted by a simulated diesel generator set model according to an exemplary embodiment.

[0021] Figure 12 illustrates the LAF output and the simulated, measured and desired speed responses of the diesel generator set 10, for a real step power input, according to an exemplary embodiment.

Figure 13 shows the LAF output with the PSO-regulated LAF parameters as described above, a desired speed response, and the speed response with and without LAF, for a real step power input, according to an exemplary embodiment.

[0023] Figure 14 is a block diagram of a computer system suitable for use with LAF systems and methods according to various embodiments.

DETAILED DESCRIPTION OF PREFERRED EMBODIMENT

The following description is by way of example only and is not intended to limit the present disclosure or the applications or uses of the disclosure.

The present system and method are applicable to a diesel generator set 10, of which a block diagram is illustrated in Figure 1. The diesel generator set 10 may include a diesel engine 12, a synchronous machine or generator 14, an exciter 16, a speed governor 18 and an automatic voltage regulator (“AVR”) 20. The diesel generator set 10 can be connected to a real power load 22.

[0026] The interaction between voltage control and speed control of diesel generator set 10 is shown in the functional block diagram of Figure 2. Each block in Figure 2 represents a Laplace operator transfer function. In the following description for Figures 2 to 3 and 6, the transfer function block (s) that are described are referenced by the part or function or effect of the physical system or software function that those blocks represent. .

The input to the functional block diagram, Vref 17, is a constant representing the desired generator voltage. The governor 18 may be operably coupled to a fuel pump 24. The output of the fuel pump block 24 represents the mechanical torque, denoted as Tm. The electrical torque, Te, can be determined based on generator voltage 15, denoted as Vt, and the resistive load, R 19. Frictional torque can be denoted as Tf. Speed governor 18 may include a communication interface for receiving a control output. Speed governor 18 may adjust the speed of diesel engine 12 based on all or part of the control output received.

The total moment of inertia 36 can be determined, in part, based on the construction of the diesel engine 12 and the generator 14, which includes the rotational parts of each, and is represented as a function of the total engine inertia. (H) 54. The AVR 20 may be operably coupled to the exciter 16 via a power amplifier 25. The desired generator set speed response 10 is represented as ref 30. The desired actual speed response is represented as ω31. A feedback signal, ί (ω) 21, can be generated as a function of Tm, Tf, Te, and H 54.

A sudden increase in actual power load from generator set 1022 causes increased load torque on diesel engine 12. Since load torque exceeds diesel engine torque and speed governor 18 cannot respond instantly the engine speed slows down. Upon detecting this deceleration, the governor 18 may instruct the fuel pump 24 to increase the fuel supplied to the diesel engine cylinders 12. Since the voltage generated is proportional to the diesel engine speed 12, the generator voltage 14 is reduced due to the induced reaction and internal voltage drops of generator 14. The AVR 20 compensates for the reduction in generator voltage 14 by increasing the field current of generator 14 supplied by exciter 16.

Since the actual power is proportional to the square of the voltage, a fast acting AVR 20 will significantly prevent diesel engine speed recovery 12 by quickly recovering the voltage and thus putting more load on the engine. Thus, a speed governor reaction 18 at a speed change due to the application of actual power load 22 is slower than that of the actual power change. In a modern AVR, a real power load measurement of 22 is available. Thus, in order to reduce the speed deviation when a large real power load 22 is applied, a load anticipation feature ("LAF") that adds an advance feed signal proportional to the change in real power load. 22 for the speed governor error sum point 18 during transient conditions, is more effective than conventional approaches that change the AVR 20 setpoint based on a specific underfrequency curve.

Particle Swarm Estimate

The Particle Swarm Improvement (“PSO”) technique can be applied to estimated programmable parameters required to deploy LAF. PSO is an evolutionary computing technique inspired by the social behavior of mass or schooling of fish.

In PSO, potential particles (solutions) disperse through the problem space following the ideal current particles. Each particle tracks its coordinates in problem space and communicates the best revealed solution for the other particles. This communication between particles allows for intelligent decision in each iteration to reveal the best possible solution for a given set of programmable parameters.

DETERMINING A SPEED CONTROL SYSTEM

EQUIVALENT LINEAR

In order to deploy the LAF, programmable parameters of an equivalent linear speed control system must be determined. A functional block diagram of the equivalent linear speed control system 26 is shown in Figure 3, according to an exemplary embodiment. Each block in Figure 3 represents a Laplace operator transfer function. Equivalent linear speed control system 26 may include quantities representing a governor, actuator 34, engine combustion function and moment of inertia 36. Engine performance is affected nonlinearly by changes in operating speed, power load. real and AVR response. Thus, in order to deploy the desired speed response of the generator speed control system including the LAF, an equivalent linear speed control system may be derived, the response of which approximately corresponds to the nonlinear speed response of the set. of current diesel generators 10.

The actual power output (PE) 28 of generator 14 and the desired speed response ωΓβί 30 are inputs to the equivalent linear speed control system 26. The equivalent linear speed control system 26 may include a type of Proportional, Integral, and Derivative (PID) controller 32 of a speed governor 18, an actuator 34, a moment of inertia 36, and Speed Drop 38 as shown in Figure 3. Controller (PID) 32 may include one or more programmable parameters . In the exemplary embodiment of Figure 3, programmable controller (PID) parameters 32 may include KP 40, Ki 42, KD 44, and TD 45. TD 45 represents a derived time constant that is a design constant for the PID. The term “s” 48 represents the Laplace operator. Actuator 34 may include an actuator time constant (TA) 50. Mechanical torque (Tm) 52 may be determined as a function of actuator output 34 and generator actual power output (PE) 28. of inertia 36 of diesel engine 12 can be represented as a function of total engine inertia (H) 54. The output of the equivalent linear speed control system 26 can be w 56, which represents the desired speed response of the diesel engine 12.

A block diagram of a procedure according to an exemplary embodiment that can be used to identify the programmable parameters of the equivalent linear speed control system 26 is illustrated in Figure 4. A maximum resistive load 58 is applied to the diesel generator set 10 and generator 14 actual power output (PE) 28 and actual speed control system response (Δω) 62 are recorded every quarter cycle for a predetermined time.

The actual power output (PE) 28 of generator 14 may also be used as an input to the equivalent linear speed control system 26, which produces an equivalent linear speed control system response (Aojm) 64. Actual speed control system response (Δω) 62 and equivalent linear speed control system response (Au> m) 64 can be used to produce an input 6 for a PSO 68 Programmable Parameter Setting Rule that You can adjust the programmable parameters of the equivalent linear speed control system 26 using the PSO technique. As described more fully below, the Adjustable Programmable Parameters 70 may be provided for the equivalent linear speed control system 26. The equivalent linear speed control system (Au) m response 64 can be determined using Programmable Parameters Adjusted 70. The above process may be repeated until the equivalent linear velocity control system response (Δω ™) 64 sufficiently corresponds to the actual velocity control system response (Δω) 62.

The PSO 68 Programmable Parameter Setting Rule may be initialized with a group of five particles (solutions) based on the particular configuration (ie equipment size, equipment ratings) of the diesel generator set 10 and then the PSO 68 Programmable Parameter Setting Rule can search through problem space for ideals by following the ideal particles revealed so far. Each particle may include one or more programmable parameters, which coincide with the programmable parameters of the equivalent linear speed control system model 26. In an exemplary embodiment, the programmable parameters may include H 54, TA 50, KP 40, Ki 42, KD 44 and Speed Drop 38.

[0037] The step response (Δω ™) 64 of the equivalent linear speed control system model 26, Oajm (/ c), can be calculated for k = 1, ..., N where N is the total number samples based on each particle and generator 14 actual power (PE) output samples 28 as described above. The step response (Aoom) 64 of the equivalent linear speed control system model 26 can be compared to the real speed control system response (Δω) 62. A sampled value of the real system response (Δω) at stage k can be denoted as Δω (/ τ). The suitability function for choosing the best particle is the sum of the square of the differences between Δω (/ (), and Δω ^ / ί), k = 1, ..., N as follows: [0038] The suitability function is evaluated for the Δω, η (Ι <) 64 that results from each particle. Each time the suitability function is evaluated, an xnself value for each particle can be determined. The resulting values xnself can be used to determine a value x9 / oi3a /, which represents the best particle of all particles.

After revealing the best values of one or more of the programmable parameters H 54, TA 50, KP 40, Ki 42, KD 44 and Speed Drop 38, each particle updates its velocity and positions with the following equations: [0040] ] In the above equations, vn is the particle velocity, xn is the current particle (solution), and x9lobal are defined as the best value for a particle and the best value among all particles, respectively, α is a weight of inertia that can represent the tendency to follow the direction each particle has moved during previous iterations. rand1 and rand2 are random numbers between 0 and 1. βί and β2 are learning factors that, together with the weight of inertia, represent the tendency of each particle to follow a new direction based on xselfn and x9lobal values.

For practical reasons, search surface limits may be limited by setting the values of one or more programmable parameters of the equivalent linear control system 26 to the values listed in Table 1 below. If the value of any one of one or more programmable parameters falls below the corresponding Minimum Value shown in Table 1, the value of that programmable parameter is set to the corresponding Minimum Value. Similarly, if the value of any one of one or more programmable parameters exceeds the corresponding Maximum Value shown in Table 1, the value of that programmable parameter is set to the corresponding Maximum Value.

TABLE 1 PARTICULAR SEARCH SPACE

A computational procedure 70 for implementing PSO Programmable Parameter Setting Rule 68 is shown in Figure 5, according to an exemplary embodiment. The steps of computational procedure 70 are now described more fully.

Step 0 <72>: Set the Iteration Counter to zero.

Step 1 <74>: Determine the actual speed control system response (Δω) 62 by applying the generator nominal actual power load 14 as the actual power load 22 and sampling the actual power output ( PE 28 of generator 14 and the response (Δω) 62 of the actual speed control system 60 every quarter cycle to a predetermined sampling interval, such as five seconds. One skilled in the art will recognize that a shorter or longer sampling interval may be used without departing from the scope of the present disclosure. Samples resulting from generator 14 actual power output (PE) 28 and actual speed control system response (w) 62 may be denoted respectively as PE (k) and Οω (/ ί), for k = 1 , ..., N, where N represents the total number of samples measured.

Step 2 <76>: Initialize each particle position based on equipment specifications (ie equipment size, equipment ratings) on diesel generator set 10 and its speed to zero.

Step 3 <78> Set the Particle Counter to zero.

Step 4 <80>: Calculate the response of the equivalent linear velocity control system model (Δω ™) 64 based on the actual generator power output 14 (Pn (k)) 28, for the particle position selected xn. The response of the equivalent linear velocity control system model (Au) m) 64 can be denoted as Au (k), for k = 1, ..., N, where N represents the total number of samples measured. Each PE sample (k) produces a corresponding Acom (/ c).

Step 5 <90>: Calculate the suitability function using equation (1) above to choose the best value xnself based on the actual speed control system response (Δω (/ τ)) 62 obtained in Step 1 <74> and response of the equivalent linear velocity control system model (Oujm (k)) 64 obtained in Step 4 <80>.

Step 6 <82>: Reveal the best particle value, update x9 '/ ote value, update particle position and velocity with equations (2) and (3) above, and as described above.

Step 7 <84>: Increment the Particle Counter.

Step 8 <86>: If the Particle Counter is equal to the Number of Particles, go to Step 9 <92>. If not, go to Step 4 <80>.

Step 9 <92>: Increment the Iteration Counter.

Step 10 <94>: If the Iteration Counter exceeds the Maximum Number of Iterations, complete computational procedure 70. If not, go to Step 3 <78>.

At the conclusion of computational procedure 70, one or more programmable parameters of the equivalent linear speed control system 26 was determined.

DEFINITION OF LAF PROGRAMMABLE PARAMETERS

Figure 6 shows a functional block diagram of an exemplary embodiment of a diesel generator set speed control system 10 modified by an LAF 96. Each block in Figure 6 represents an operator-shaped transfer function. from Laplace. The LAF 96 is a controller modifier for the governor that becomes enabled when the rate of change in actual power load 22 is greater than a programmable amount. The programmable amount can be determined based on the speed response of the diesel generator set 10 for actual power load variations 22 with the LAF disabled. Typically, the programmable amount may be 80% of the actual power rate of generator 14, although other higher or lower values may be used without departing from the scope of the present disclosure. Lower programmable threshold values can be used if LAF 96 is to be active for lower percentages of the actual generator power rate 14.

LAF 96 adds a forward feed signal 106 proportional to the actual power load change 22 to the speed governor error sum point 18 during transient conditions. The rate of change in actual power load 22 may be obtained with a high pass filter 108. The phase delay due to speed governor 18 and actuator 34 may be compensated with an earlier phase filter 110. A zone block Inoperative 112, which may have an input from the high pass filter 108 or the previous phase filter 110, may be used to make the LAF 96 active only if the rate of change in actual power load 22 exceeds a higher programmable limit. Preferably, the dead zone block 112 disables LAF 96 when the absolute value of its input falls below a programmable limit. The programmable deadband block limit 112 may typically be 10% of the actual power rate of generator 14 per second. One skilled in the art will recognize that other values, which may or may not depend on the actual power rate of generator 14, may be used without departing from the scope of the present disclosure. LAF 96 may include an enabled block 114 which can be used to disable LAF 96 based on an enabled input 118 (ΕΝ). The LAF 96 may also include a signal amplifier 120, which has an output gain 100 (KLa). The exemplary embodiment of Figure 6 may be used to estimate one or more programmable parameters of LAF 96. The one or More programmable parameters of LAF 96 may include the removal time constant 98 (TLa), output gain 100 (KLa), phase advance constant 102 (TLd), and delay time constant 104 (TLg) - The ability of the LAF 96 to achieve the desired speed response of the equivalent linear speed control system 26 depends on the values of these programmable parameters. PSO can be used to reveal a set of appropriate LAF 96 programmable parameters. The equivalent linear speed control system 26 identified above is used to reveal a set of optimized LAF 96 programmable parameters.

[0058] A block diagram works! of a procedure according to an exemplary embodiment which can be used to identify the programmable parameters of the LAF 96 is illustrated in Figure 7. A maximum resistive load 122 is applied to the diesel generator set 10 and the actual power output ( PE 28 of generator 14 is recorded every quarter cycle for a predetermined amount of time. The actual power output (PE) 28 of generator 14 can be used as an input to the LAF and equivalent model 26 linear control system. The desired speed response 124 can be determined based on the actual speed measurement (Δω). and the scaling factor (Desired) using the following equation: (4) The equivalent linear speed control system response (AüJm) 64 can be determined based on the actual power output (PE) 28 14. Desired speed response (Αωά) 124 and equivalent linear speed control system response (Or) m) 64 can be used to produce an input 126 for a PSO LAF Programmable Parameter Setting Rule 128 which can adjust the LAF 96 programmable parameters using the PSO technique. As described more fully below, Adjusted Programmable Parameters 130 may be provided to the equivalent linear speed control system 26. The equivalent linear speed control (Oojm) response 64 may be determined using Adjusted Programmable Parameters 130. The above process may be repeated until the equivalent linear velocity control system response (Au) m) 64 sufficiently corresponds to the desired velocity response (Aood) 124. The output of LAF 96, VLAF 132 can be determined with based on actual power output (PE) 28 from generator 14 and Adjustable Programmable Parameters 130.

The LAF Programmable Parameter Setting Rule by PSO 128 can be initialized with a group of five particles (solutions) with randomly selected particle values within the search space. The LAF Programmable Parameter Setting Rule by PSO 128 then searches through the ideal search space following the ideal particles revealed so far. The equivalent linear velocity control system response (A (jjm) 64) can be calculated based on the equivalent linear velocity control system 26, each particle and samples from generator actual power output (PE) 28. calculated from the equivalent linear velocity control system response (Aojm) 64 can be denoted as Auom (k) A desired value of the desired velocity response (A (jjd) 124 at stage k can be denoted as Aqjc / (/ c) .The resulting calculated values can be compared to the desired velocity response Awd (/ c) The suitability function for choosing the best particle is the sum of the squared differences between Aojc / (k) and Δω ™ (/ ί) , k = 1, ..., N as follows: (5) [0061] After revealing the best values of one or more programmable parameters of LAF 96, TLA 98, KLA 100, TLD 102, TLG 104, each particle updates your velocity and positions with the following equations: [0062] In the above equations, vn is the particle velocity, xn is the current particle (solution), and x9lobal are defined as the best value for a particle and best value among all particles, respectively, α is an inertial weight that may represent the tendency to follow the direction that each particle moved during previous iterations. randi and rand2 are random numbers between 0 and 1. βί and β2 are learning factors, which together with the weight of inertia represent the tendency of each particle to move in a new direction based on xselfn and xglobal values. that the values α, β1, and β2 used in determining one or more LAF 96 programmable parameters differ from those used for determining the programmable parameters of the equivalent linear speed control system 26 described above.

[0063] For practical reasons, search surface boundaries may be limited by setting the values of one or more LAF 96 programmable parameters to the values listed in Table 2 below. If the value of any one of one or more programmable parameters falls below the corresponding Minimum Value shown in Table 2, the value of that programmable parameter is set to the corresponding Minimum Value. Similarly, if the value of any one of one or more programmable parameters exceeds the corresponding Maximum Value shown in Table 2, the value of that programmable parameter is set to the corresponding Maximum Value.

TABLE 2 Particle Search Space

A computational procedure 134 for implementing the LAF Programmable Parameter Setting Rule by PSO 128 is shown in Figure 8, according to an exemplary embodiment. The steps of computational procedure 134 are now described more fully.

Step 0 <136>: Set the Iteration Counter to zero.

Step 1 <138>: Record each of the calculated values every quarter of a cycle for a predetermined sampling interval, such as five seconds, using the above equation (4) and calculate the desired velocity response (Δω ^ 124. One of ordinary skill in the art will recognize that a shorter or longer sampling interval may be used without departing from the scope of the present disclosure. Samples resulting from the desired velocity response Aud may be denoted as Δα) d (k), for k = 1, ..., N, where N represents the total number of recorded samples.

Step 2 <140>: Initialize each particle position and its velocity to zero.

Step 3 <142> Set the Particle Counter to zero.

Step 4 <144>: Calculate the response of the equivalent linear velocity control system (Awm) 64 model based on the actual generator power output 14 (Pe (7c)) 28, for the selected particle position xn. The response of the equivalent linear velocity control system (Tuna) 64 model can be denoted as Aajm (k), for k = 1, ..., N, where N represents the total number of samples measured. Each sample of Pe (7 () produces a corresponding Aajm (/ r).

Step 5 <146>: Calculate the suitability function, equation (5), to choose the best R value based on the desired velocity response (Aüüd) 124 obtained in Step 1 <138> and the model response. equivalent linear speed control system (A6um) 64 obtained in Step 3 <142>.

Step 6 <152>: Update particle position and velocity with equations (6) and (7) above and update the x9lobal value as described above.

Step 7 <148>: Incrementing the Particle Counter.

Step 8 <150>: If the Particle Counter is equal to the Number of Particles, go to Step 9 <154>. If not, go to Step 4 <144>.

Step 9 <154>: Increment the Iteration Counter.

Step 10 <156>: If the Iteration Counter exceeds the Maximum Number of Iterations, complete computational procedure 134. If not, go to Step 3 <142>.

In describing elements or features and / or modalities thereof, the articles "one", "one", "o", "a", "said" and "said" are intended to mean that there is one or more among the elements or features. The terms "comprising", "including" and "having" are intended to be inclusive and mean that there may be additional elements or features beyond those specifically described.

Those skilled in the art will recognize that various changes may be made to the exemplary embodiments and implementations described above without departing from the scope of the disclosure. Accordingly, all material contained in the above description or shown in the accompanying drawings should be construed as illustrative and not in a limiting sense.

It is further to be understood that the processes or steps described herein should not be construed as necessarily requiring their performance in the particular order discussed or illustrated. It should further be understood that alternative or additional processes or steps may be employed.

IMPLEMENTATION OF LAF SYSTEMS AND METHODS

In order to show its effectiveness and applicability to the industrial environment, the system and method were deployed on a commercially available AVR designed for a small (less than 10MVA) generator set. In general, this type of regulator has been designed with great cost effectiveness and memory size and computing power are limited.

The AVR consists of a microprocessor and signal conditioning circuits for generator voltage, current, and PWM controlled regulator output. Generator voltage and current are sampled at 12 bits resolution after anti-aliasing filters. The generator voltage rms calculation is calculated every quarter cycle (4.16 msec for 60Hz system).

[0081] In order to provide a self-regulating AVR simple user interface, a software application has been developed. The following features are implemented in the software application: [0082] 1. Basic diagnostic functions (wiring validation and measurement calibration).

[0083] 2. Ability to record actual generator speed and power responses over a fixed period of time when the actual power load is applied.

[0084] 3. Ability to identify an equivalent speed control system using the PSO technique.

4. Ability to generate the desired velocity response Owd.

[0086] 5. Ability to estimate the LAF parameters to match the equivalent LAF system speed for the desired velocity response obtained in step 4.

[0087] 6. Ability to set LAF parameters and check speed response is approximately equal to desired response.

[0088] A field engineer can enable software application auto-tuning mode and the LAF parameters can be calculated to a desired value of K desired. [0089] A Real-Time Digital Simulator (RTDS) was used to verify performance. system and method disclosed above. A fully detailed generator model in the RTDS library can be used as part of the performance check. A representation of the exciter-rectifier motor and exciter motor is shown as functional block diagrams in Figures 9A and 9B. An AVR into which the previously disclosed system and method were deployed has also been connected to an RTDS.

RTDS, which includes one embodiment of the LAF systems and methods described above, may be deployed using MATLAB® software to simulate LAF systems and methods, and BESTCOMPLUS® software may be used to communicate with a device. to get required test results. One of ordinary skill in the art will appreciate that other suitable means may be used to deploy RTDS which includes, but is not limited to, dedicated developed software that runs on commercially available operating systems and hardware and other suitable communication software.

[0091] Figure 10 shows a real recorded speed response for a particular applied real power input. Actual generator power and speed response are shown as values per unit. Based on these responses, an equivalent linear speed control system was identified as described above. Estimated LAF programmable parameter values of the identified equivalent linear speed control system are shown in Table 3.

TABLE 3 LAF PARAMETERS ESTIMATE

[0092] Estimated LAF programmable parameter values are slightly different from Model parameter values. The difference may result from nonlinear functions present in alternator-rectifier motor and exciter RTDS models. Figure 11 shows the actual recorded speed response of Figure 10 and the equivalent linear control system speed response using the Estimated parameters shown in Figure 3. Despite the differences between the Model and Estimated LAF programmable parameter values , the equivalent linear speed control system response time is very close to the actual recorded speed response as illustrated in Figure 11.

TABLE 4 LAF PARAMETERS ESTIMATE

[0093] Figure 12 illustrates the LAF output and the simulated, measured and desired speed responses of the diesel generator set 10 for a real step power input are shown. The measured velocity response, f measured, shown is the actual recorded velocity response shown in Figures 10 to 11. The desired velocity response, f desired, can be obtained by multiplying the measured velocity response, f, by Desired K = 0.5 or otherwise suitable for producing a speed deviation that meets or exceeds the requirements of IS08528-5. RTDS, with LAF enabled, can be used to determine simulated system speed response, f simulation. The LAF output is also shown in magenta. For simulation purposes, the LAF output can be calculated from the actual recorded power that was obtained with inactive LAF. In this case, the actual response with active LAF will be slightly different as the speed response is affected by LAF. As can be seen from Figure 12, the velocity response with the activated LAF, f simulation, closely approximates the desired velocity response, f desired.

Figure 13 shows the effect of regulating LAF programmable parameters using the PSO method as described above. The LAF output, the desired speed response, the desired f, and the speed response with and without the LAF, for a real step power input are shown. As can be seen from Figures 12 to 13, the velocity response with the set LAF parameters, f with LAF, shown in Figure 13 is even closer to the desired velocity response, f than the desired response. velocity with the LAF using parameters before setting, f simulation, from Figure 12.

[0095] In an illustrative embodiment, the LAF and PSO tuning method were implemented using a typical Diesel Generator Set, which includes a 125 kVA, 208 Vac, 1,800 RPM 3φ synchronous generator. No load excitation for this generator was provided by a 0.3 Amp, 7 Vdc, ac self-excited exciter.

A real power load of 100 kW has been applied to the Diesel Generator Set. In order to reveal an appropriate set of LAF parameters, the diesel engine speed deviation was recorded in response to the applied load. An equivalent speed system has been identified as described above. Estimated programmable parameter values resulting from equivalent linear velocity system parameters are shown in Table 5.

TABLE 5: ESTIMATE OF EQUIVALENT LINEAR SPEED CONTROL SYSTEM

Based on the identified equivalent speed system parameters, the LAF parameters are calculated and shown in Table 6.

TABLE 6. ESTIMATE OF LAF PROGRAMMABLE PARAMETERS

Similar results to those obtained with RTDS were obtained using the Diesel Generating Set. Diesel Generator Set speed deviation without the LAF exceeds the speed requirement of IS085285-2 from a 10% speed drop. With LAF enabled, your desired speed deviation with Desired K = 0.5 is easily achieved, resulting in a speed deviation well within the requirements of IS085285-2.

COMPUTER OPERATING ENVIRONMENT

Referring to Figure 14, an operating environment for an illustrated embodiment of a system and / or method for an AVR or speed governor including LAF systems and methods as described herein is a computer system 1000. with a computer 1002 comprising at least one high speed central processing unit (CPU) 1004, together with a memory system 1006 interconnected to at least one bus structure 1008, an input device 1010 and an output device 1012 These elements are interconnected by at least one bus structure 1008. In an alternative embodiment, LAF systems and methods may be deployed on a standalone device or incorporated into another component in the system that provides the operating environment described above.

The computer 1002 may include an input data interface for receiving a measurement of the actual power load applied to the diesel generator set and a communication interface that may be operably coupled to the governor communication interface. 18. The communication interface of the computer 1002 may transmit a control output to the speed governor communication interface 18.

As contemplated above, the input and output devices may include a communication interface that includes a graphical user interface. Any or all of the network interface computer components and communication methods and systems may be any communication device that includes, but is not limited to, a laptop, PDA, mobile phone, as well as potentially a device. dedicated. The software may be deployed as any app in it and still within the scope of this disclosure.

[00102] The illustrated CPU 1004 is a family design and includes a 1014 arithmetic logic unit (ALU) for performing calculations, a 1016 record collection for temporary data and instruction storage, and a 1018 control unit for controlling system operation. Any of a variety of microprocessors are equally preferred, but not limited to, for CPU 1004. This illustrated embodiment operates on an operating system designed to be portable to any of these processing platforms.

Memory system 1006 generally includes high speed main memory 1020 in the form of a medium, such as random access memory (RAM) and read-only memory (ROM) semiconductor devices that are typical in a medium non-transient computer writable. The present disclosure is not limited thereto and may also include secondary storage 1022 in the form of long term storage media such as floppy disks, hard disks, tape, CD-ROM, flash memory, etc., and other devices that store data with the use of electric, magnetic and optical media or other recording media. The main memory 1020 may also include, in some embodiments, a video display memory for displaying images through a display device (not shown). Those skilled in the art will recognize that memory system 1006 may comprise a variety of alternative components that have a variety of storage capacities.

Where applicable, an input device 1010, and an output device 1012 may also be provided in the system as described herein or embodiments thereof. The input device 1010 may comprise any keyboard, mouse, physical transducer (e.g. a microphone), and may be interconnected to the computer 1002 via an input interface 1024, such as a graphical user interface, associated with or separate from the interface. described above that includes the antenna interface for wireless communication. Output device 1012 may include a display, a printer, a transducer (e.g., a speaker), etc., and may be interconnected to computer 1002 via an output interface 1026 which may include the communication interface described above. which includes the antenna interface. Some devices, such as a network adapter or modem, can be used as input and / or output devices.

As familiar to those skilled in the art, computer system 1000 additionally includes an operating system and at least one application program. The operating system is the software suite that controls the operation of the computer system and the allocation of fonts. . The application program is the software suite that performs a task desired by the LAF system and method and or any of the processes or process steps described above using computer sources made available through the operating system.

In accordance with the practices of persons skilled in the art of computer programming, the present disclosure is described below with reference to symbolic representations of operations that are performed by the computer system 1000. Such operations are sometimes referred to as executed by computer. It will be appreciated that operations that are symbolically represented include manipulation by the CPU 1004 of electrical signals representing data bits and maintenance of data bits at memory locations in memory system 1006, as well as other signal processing. Memory locations where data bits are kept are physical locations that have particular electrical, magnetic, or optical properties that correspond to the data bits. One or more embodiments may be tangibly implemented in a program or programs defined by computer executable instructions which may be stored in a computer readable medium. The computer readable medium may be any of the devices or a combination of the devices described above in connection with memory system 1006.

The foregoing disclosure thus discloses multiple systems and methods that may be produced from various elements and steps which may or may not be present in any particular system or method to be used at a particular time or in a particular definition and method. It thus reveals many permutations of systems and methods. Earlier disclosure allows for variation and selection of features, elements, and steps that depend on the user or users.

Claims (26)

1. Method for providing a load anticipation feature for a diesel generator set that includes a diesel engine, a generator, a speed governor, and an automatic voltage regulator CHARACTERIZED by the fact that it comprises the steps of: determining a equivalent linear speed control system, where the linear speed control system response corresponds approximately to the nonlinear speed response of the diesel generator set; and determining at least one programmable load anticipation feature parameter based on a real power load applied to the diesel generator set and a desired speed response from the linear speed control system. Method according to claim 1, characterized in that the step of determining an equivalent linear speed control system further comprises the step of determining a programmable parameter set of the linear speed control system using particle swarm enhancement means the set of programmable parameters determined by a predetermined number of particles, wherein the position of each particle represents a different set of programmable parameters; and for each particle: a. initialize each particle position; B. determining an actual value of the diesel generator set speed response based on a real generator power load; ç. calculate a linear velocity control system response with a selected particle position using the actual power load and velocity offset; d. determining a particle suitability assessment to reveal a better eigenvalue and an overall better value by comparing the calculated response to the actual velocity response value; and is. update a position and a velocity for each particle based on the best determined eigenvalue and overall best value. A method according to claim 1, characterized in that the step of determining a linear speed control system further comprises the steps of: determining a programmable parameter set of the equivalent linear speed control system using particle swarm enhancement, wherein the programmable parameter set is determined by a fixed number of particles, wherein the position of each particle represents a different set of programmable parameters; and for each set of programmable parameters: Step 0: Initialize Change counter = 0; Step 1: Determine a real speed control system, Acu (k), for the diesel engine using the generator's actual power output, Pe (k) as a predetermined number of samples, k = 1. N, for a predetermined period of time; Step 2: Initialize each particle position to typical values based on diesel genset equipment specifications and each particle speed to zero; Step 3: Initialize Particle Counter = 0; Step 4: Calculate the equivalent linear velocity control system response, A6um (k), with a selected particle position, based on the actual power output, Ρε (^), k = 1, ..., N; Step 5: Calculate a suitability function: v ';' 11 to choose the best value fl based on the actual speed control system response and the response of the equivalent linear speed control system model; Step 6: Reveal the best particle value 11, update gfobal. value "and update a new particle position and velocity using the equations: Step 7: Until all particles are calculated based on Particle Counter and predetermined number of particles, increment the Particle Counter and repeat steps 4 to 6 above for each particle Step 8: If the Particle Counter is equal to the total Number of Particles in the programmable parameter set, go to Step 9, If not, go to Step 4; Step 9: Increment the Particle Counter Iteration and Step 10: If the Iteration Counter is less than or equal to a predetermined Maximum Number of Iterations, go to Step 3. Method according to claim 2, characterized in that the programmable parameter set of the equivalent linear speed control system comprises at least one of: H = total motor inertia, TA = actuator time constant, KP, K ?, KD = (PID) controller parameters and Speed Drop. Method according to claim 3, characterized in that the programmable parameter set of the equivalent linear speed control system comprises at least one of: H = total motor inertia, TA = actuator time constant, KP, K7, KD = (PID) controller parameters and Speed Drop. A method according to claim 1, characterized in that the step of determining at least one programmable load anticipation feature further comprises the step of determining a programmable load anticipation feature set with the use of particle swarm enhancement, wherein the programmable parameter set is determined by a predetermined number of particles; and for each particle: a. initialize each particle position; B. determining a desired speed response from the equivalent linear speed control system based on the nominal actual generator power load; ç. calculate a velocity response of the linear velocity control system and the charge anticipation feature with a particle position selected using actual power load and velocity deviation; d. determining a particle suitability assessment to reveal a better eigenvalue and an overall better value by comparing the calculated velocity response with the desired velocity response; and is. update a position and a velocity for each particle based on the best determined eigenvalue and overall best value. A method according to claim 1, characterized in that the step of determining at least one programmable load anticipation feature further comprises the step of: determining a programmable parameter set of the load anticipation feature with the use of particle swarm enhancement, wherein the programmable parameter set is determined by a fixed number of particles, wherein the position of each particle represents a different programmable parameter set; and for each set of programmable parameters: Step 0: Initialize Change Counter = 0; Step 1: Determine a desired speed response, Aa> d (k), from the linear speed control system using the actual power output, PE (k), as a predetermined number of samples, k = 1 ,, N, for a predetermined period of time; Step 2: Initialize each particle position and each particle velocity to zero; Step 3: Initialize Particle Counter = 0; Step 4: Calculate the response of the equivalent linear velocity control system and the load anticipation feature, Ao> m (k), with a selected particle position, based on actual power output, PE (k), / f = 1. N; Step 5: Calculate a Suitability Function: To choose the best value "based on the equivalent linear speed control system, load anticipation feature response, and speed response ^ Desired; x" if Step 6: Reveal the best particle value, update the x9lobal value and update a new particle position and velocity using the equations: Step 7: until all particles are calculated based on the Particle Counter and predetermined number of particles, increment the Particle Counter and repeat steps 4 through 6 above for each particle Step 8: If Particle Counter equals Total Number of Particles in programmable parameter set, go to Step 9, If not, go to Step 4 Step 9: Increment the Iteration Counter and Step 10: If the Iteration Counter is less than or equal to a predetermined Maximum Number of Iterations, go to Step 3. Method according to claim 1, characterized in that the desired velocity response, Au) d (k), of the equivalent linear velocity control system is determined based on the desired K and the actual velocity response of the diesel genset. A method according to claim 6, characterized in that the programmable load anticipation feature set comprises at least one of: Tw = removal time constant, Kw = output gain, TAP = constant phase advance and TRE delay = delay time constant. Method according to claim 7, characterized in that the programmable load anticipation feature set comprises at least one of: Tw = removal time constant, Kw = output gain, TAVA = constant Phase Advance and Delay = Delay Time Constant. A method according to claim 1, characterized in that it further comprises the steps of: measuring a real power load applied to the diesel generator set by a sensor electrically coupled to the diesel generator set; and providing at least one programmable load anticipation feature to the computer data processing apparatus having computer memory, wherein the computer memory contains computer executable instructions for deploying the linear speed control system and the feature. load anticipation, the computer executable instructions for deploying the load anticipation feature that includes computer executable instructions for generating a control output based on the actual measured power load and at least one programmable load anticipation parameter charge; Provide the control output to the governor, where the governor adjusts the speed of the diesel engine based on the control output. A method according to claim 11, characterized in that the step of providing control output to the speed governor further comprises the step of blocking control output to the speed governor unless the actual power load is present. exceeds a predetermined percentage of the generator rated power load. A method according to claim 11, characterized in that the step of providing control output to the speed governor further comprises the step of enabling control output to the speed governor when the rate of load change actual power exceeds a programmable amount. A system for providing a load anticipation feature for a diesel generator set comprising a diesel engine, a generator, a speed governor and an automatic voltage regulator comprising: a computer data processing apparatus having a communication interface and is operably coupled to computer memory containing computer executable instructions, wherein the computer data processing apparatus is for executing computer executable instructions, wherein computer executable instructions are structured to: determining a linear speed control system, wherein the linear speed control system response corresponds approximately to the nonlinear speed response of the diesel generator set; and determining at least one programmable load anticipation feature parameter based on a real power load applied to the diese generator set! and a desired velocity response of the linear velocity control system; deploy linear speed control system and load anticipation feature. A system according to claim 14, characterized in that it further comprises: a sensor for measuring the actual power load applied to the diesel generator set; an input data interface for receiving a measurement of the actual power load applied to the diesel generator set; wherein the computer executable instructions are further structured to generate a control output based on the actual measured power load and at least one programmable load anticipation parameter; and wherein the speed governor includes a communication interface operably coupled to the communication interface of the computer data processing apparatus, wherein the speed governor communication interface receives control output transmitted from the communication interface. of a computer data processor apparatus wherein the speed governor adjusts the speed of the diesel engine based on the control output. A system according to claim 14, characterized in that the computer executable instructions structured to determine a linear speed control system are additionally structured to: determine a programmable parameter set of the linear speed control system with the use of particle swarm enhancement, wherein the programmable parameter set is determined by a predetermined number of particles, wherein the position of each particle represents a different programmable parameter set; and for each particle: a. initialize each particle position; B. determining an actual value of the diesel generator set speed response based on a real generator power load; ç. calculate a linear velocity control system response with a selected particle position using actual power load and velocity deviation; d. determining a particle suitability assessment to reveal a better eigenvalue and an overall better value by comparing the calculated response to the actual velocity response value; and is. update a position and velocity for each particle based on the best eigenvalue determined and the overall best value. A system according to claim 14, characterized in that the computer executable instructions are structured to determine a linear velocity system are further structured to: determine a programmable parameter set of the linear velocity control system with the use of particle swarm enhancement, wherein the programmable parameter set is determined by a fixed number of particles, wherein the position of each particle represents a different programmable parameter set; and for each set of programmable parameters: Step 0: Initialize Change counter = 0; Step 1: Determine a real speed control system, Δω (Ι <), for the diesel engine using the generator's real power output, Pe {k) as a predetermined number of samples, k = 1, ..., N, for a predetermined period of time; Step 2: Initialize each particle position to typical values based on diesel genset equipment specifications and each particle speed to zero; Step 3: Initialize Particle Counter = 0; Step 4: Calculate the equivalent linear velocity control system response, Aoom (k), with a selected particle position, based on the actual power output, Pf (k), k = 1, ..., N; Step 5: Calculate a suitability function: to choose the best value * based on the actual speed control system response and the response of the equivalent linear speed control system model; Step 6: Reveal the best particle value, update the xgiobai value, and act on a new particle position and velocity using the equations: Step 7: Until all particles are calculated based on the Particle Counter and number particle counter, increment the Particle Counter and repeat steps 4 through 6 above for each particle; Step 8: If Particle Counter equals total Particle Number in programmable parameter set, go to Step 9, If not, go to Step 4; Step 9: Increment the Iteration Counter; and Step 10: If the Iteration Counter is less than or equal to a predetermined Maximum Iteration Number, go to Step 3. System according to claim 16, characterized in that the programmable parameter set of the equivalent linear speed control system comprises at least one of: H = total motor inertia, TA = actuator time constant, KP, K ?, KD = (PID) controller parameters and Speed Drop. A system according to claim 17, characterized in that the programmable parameter set of the equivalent linear speed control system comprises at least one of: H = total motor inertia, TA = actuator time constant, KP, K ?, KD = (PID) controller parameters and Speed Drop. A system according to claim 14, characterized in that the computer executable instructions structured to determine at least one programmable load anticipation feature are further structured to: determine a set of anticipate feature programmable parameters loading with the use of particle swarm enhancement, wherein the programmable parameter set is determined by a predetermined number of particles; and for each particle: a. initialize each particle position; B. determining a desired speed response from the equivalent linear speed control system based on the nominal actual generator power load; ç. calculate a velocity response of the linear velocity control system and the charge anticipation feature with a particle position selected using actual power load and velocity deviation; d. determining a particle suitability assessment to reveal a better eigenvalue and an overall better value by comparing the calculated velocity response to the desired velocity response; and is. update a position and velocity for each particle based on the best eigenvalue determined and the overall best value. System according to claim 14, characterized in that the computer executable instructions structured to determine at least one programmable load anticipation feature are additionally structured to: determine a set of anticipate feature programmable parameters loading with the use of particle swarm enhancement, wherein the programmable parameter set is determined by a fixed number of particles, wherein the position of each particle represents a different programmable parameter set; and for each set of programmable parameters: Step 0: Initialize Change Counter = 0; Step 1: Determine a desired velocity response, Ao> d (k), of the linear velocity control system using the actual power output, PE (k), as a predetermined number of samples, k = 1 ,. N, for a predetermined period of time; Step 2: Initialize each particle position and each particle velocity to zero; Step 3: Initialize Particle Counter = 0; Step 4: Calculate the response of the equivalent linear speed control system and the load anticipation feature, Aium (k), with a selected particle position, based on actual power output, PE (k), k = 1 , ..., N; Step 5: Calculate a Suitability Function: To choose the best value '' based on the equivalent linear speed control system, load anticipation response, and speed responseH <= Desired; jstf # 'w' Step 6: Reveal the best particle value °, update the xglobal value, and update a new particle position and velocity using the equations: Step 7: Until all particles are calculated based on the Particle Counter and the predetermined number of Increment the Particle Counter and repeat steps 4 through 6 above for each particle Step 8: If the Particle Counter equals the total Particle Number in the programmable parameter set, go to Step 9, If not, go to Step 4: Step 9: Increment the Iteration Counter, and Step 10: If the Iteration Counter is less than or equal to a predetermined Maximum Iteration Count, go to Step 3. System according to Claim 14, characterized in that the desired velocity response, AtjOd (k), of the equivalent linear velocity control system is determined based on the desired K and the actual velocity response of the genset of diesel. A system according to claim 20, characterized in that the programmable load anticipation feature set comprises at least one of: Tw = removal time constant, Kw = output gain, Balance = constant Phase Advance and Delay = Delay Time Constant. A system according to claim 21, characterized in that the programmable load anticipation feature set comprises at least one of: Tw = removal time constant, Kw = output gain, Balance = constant Phase Advance and Delay = Delay Time Constant. A system according to claim 15, characterized in that computer-executable instructions that are further structured to provide control output to the speed governor further comprise the step of blocking control output to the speed governor. unless the actual measured power load exceeds a predetermined percentage of the generator rated power load. System according to claim 15, characterized in that the computer executable instructions are further structured to enable the transmission of the control output to the speed governor when the actual power load change rate applied to the Diesel generators exceed a programmable amount.