Method for overcoming reverse adjustment dead zone of fuel gas adjusting valve through big data statistics

Abstract

The invention discloses a method for overcoming a reverse adjustment dead zone of a gas adjusting valve through big data statistics. The method comprises the following steps that the action direction of the valve is pre-judged; according to the expected action direction of the valve, the valve adjusting instruction is forcibly switched to the average opening degree within the past 10 min time range to wait for a new valve adjusting instruction; calculating an average opening instruction of the valve within 10 minutes in real time by using a big data statistical method; according to prejudgment of the action direction of the valve, the valve adjusting instruction is limited to the average opening instruction, and a new valve PID adjusting instruction is waited again; and calculating the average ratio of the gas flow to the load within 60 minutes in real time by using a big data statistical method. The method for overcoming the reverse adjustment dead zone of the fuel gas adjusting valve through big data statistics has the advantage of overcoming the reverse adjustment dead zone of the fuel gas adjusting valve.

Description

Technical Field

[0001] The present invention belongs to the technical field of gas regulation, and in particular relates to a method for overcoming a reverse regulation dead zone of a gas regulating valve by utilizing big data statistics. Background Art

[0002] The regulating valve has a dead zone for regulation and a dead zone for reverse regulation. Whether it is automatic regulation or manual regulation, the valve action will take effect only after the dead zone for regulation is overcome. In manual regulation, the known dead zone can be quickly reversed or closed to overcome it; however, automatic regulation is a relatively smooth process. When the dead zone for regulation, especially the dead zone for reverse regulation, is large, the reverse regulation dead zone cannot be effectively or quickly overcome because there is a process for proportion and integration, resulting in poor control quality and slow response to control parameters; due to the existence of the reverse regulation dead zone, the gas flow rate will not change when the valve is opened at a certain degree. The gas flow rate will only change after the dead zone for regulation is overcome. Due to the large change in opening, overregulation will also occur.

[0003] At present, there is no method to overcome the dead zone of valve adjustment in the DCS system control system at home and abroad. When the dead zone is small, it can only be overcome by process. When the dead zone is large, it cannot be put into automatic control or it is difficult to put into automatic control. Even if automatic control is put into use, it is easy to oscillate and the effect is very poor. There is also a method of directly overriding a threshold on the basis of the current opening instruction to overcome the reverse adjustment dead zone. Due to the compressible characteristics of gas, the reverse adjustment dead zone of gas regulation is very obvious. The reverse dead zone varies with the size of the gas pipeline diameter. When the pipe diameter exceeds 1 meter, the reverse dead zone reaches about 6% of the opening. Even if a threshold is directly overridden on the basis of the current opening, the effect is general, it is easy to oscillate and over-adjust, the response is delayed, the valve moves frequently, and the life will be reduced. New methods are needed to overcome it. Summary of the invention

[0004] The purpose of the present invention is to provide a method for overcoming the reverse regulation dead zone of a gas regulating valve by utilizing big data statistics, so as to solve the above-mentioned problems existing in the prior art.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions:

[0006] A method for overcoming the reverse regulation dead zone of a gas regulating valve by using big data statistics comprises the following steps:

[0007] Predict the direction of valve movement;

[0008] According to the expected movement direction of the valve, the valve adjustment command is forced to switch to the average opening of the previous 10-minute time range and wait for new valve adjustment commands.

[0009] Using big data statistical methods, the average opening instruction of the valve within 10-30 minutes is calculated in real time;

[0010] According to the prediction of the valve movement direction, the valve adjustment command is limited to the average opening command to wait for the new valve adjustment command, and the valve is allowed to wait for 0.5 seconds at the new opening position to judge the new valve PID adjustment command; avoid over-adjustment, excessive switching and oscillation, overcome the problems of slow callback and lag, and thus overcome the reverse adjustment dead zone of the valve. Using big data statistical methods, the average valve opening command of the valve opening in the reverse time (10 minutes) is calculated in real time. When the load and gas flow change direction are predicted to determine whether reverse adjustment is required, the valve opening command is directly preset and tracked to the average opening command calculated previously, and a reverse dead zone is superimposed, so that the valve waits for new commands at the new position, thereby reducing and effectively overcoming the reverse adjustment dead zone of the valve.

[0011] Using big data statistical methods, the average ratio of gas flow to load within 60 minutes is calculated in real time;

[0012] After the load target is determined, the gas flow required for the current load target is obtained by multiplying the average ratio of the statistical gas flow to the load by the load target, and then the gas flow target required for the current load can be obtained;

[0013] The expected valve opening can be obtained based on the gas flow target and the flow characteristics of the valve;

[0014] According to the reverse regulation dead zone of the gas regulating valve, the opening range of the gas valve can be determined, and the valve instruction can be directly limited to this range, thereby avoiding excessive switching of the regulating valve and over-regulation of the gas flow.

[0015] By using big data statistical methods, the average relationship between the load and gas flow (average coal consumption) in the past hour is calculated in real time. After the load target is determined, multiplied by the statistical gas flow, it is the gas flow target required by the current load. The valve opening can be known based on the gas flow target and the flow characteristics of the valve. Then, based on the reverse adjustment dead zone of the gas regulating valve, the opening range of the gas valve can be determined (the upper and lower limits of the opening, which change dynamically with the load and coal consumption), thereby avoiding excessive switching of the regulating valve and over-adjustment of the gas flow.

[0016] Furthermore, the method for predicting the valve movement direction comprises the following steps:

[0017] Adjust the valve opening command according to the deviation between the actual load of the unit and the load target;

[0018] The valve movement direction is determined based on the average weighted change rate of the opening command in the past 3-5s. If the average weighted change rate is greater than 0, the opening command becomes larger; if the average weighted change rate is less than 0, the opening command becomes smaller.

[0019] Furthermore, the big data statistical method includes an average opening instruction statistical method and an average ratio statistical method of gas flow and load.

[0020] Furthermore, the average opening instruction statistical method comprises the following steps:

[0021] Determine whether the signal quality of the gas flow I1 is normal. If the signal quality is abnormal, the gas flow I1 data statistics are not included. If the signal quality is normal, the gas valve opening I2 data statistics and learning are performed to obtain the initial statistical value Y1.

[0022] Calculate and obtain the average opening O1;

[0023] Statistics and calculation of the average opening Y of the previous cycle, Y = O1;

[0024] After this cycle statistics are completed, the next cycle statistics will be repeated.

[0025] Furthermore, the average opening O1=(RT*120*Y+M*Y1) / (RT*120+M);

[0026] RT stands for reverse time, which represents the reverse time range to be counted for the current time and is set to 10 minutes;

[0027] M represents the weight of the current ratio, and the default value is 1. When M is larger, the weight of the current ratio is larger, and the statistics will quickly approach the current ratio.

[0028] 120 means the calculation cycle is 500ms, and the calculation is performed 120 times in 1 minute;

[0029] Y represents the memorized, statistically calculated average opening value of the previous period.

[0030] Furthermore, the statistical method of the average ratio of gas flow to load includes the following steps:

[0031] Determine whether the signal quality of the gas flow I1 and the load I3 is normal. If the signal quality is abnormal, the data statistics of the gas flow I1 and the load I3 are not included. If the signal quality is normal, the data statistics and learning of the gas flow I1 and the load I3 are performed to obtain the initial statistical value Y2, that is, the normal valve opening value I1 / I3;

[0032] The average ratio O2 was calculated;

[0033] Statistics and calculation of the average ratio Y0 of the previous period, Y0 = O2;

[0034] After this cycle statistics are completed, the next cycle statistics will be repeated.

[0035] Further, the average ratio O2=(RT*120*Y0+M0*Y2) / (RT*120+M0);

[0036] RT stands for reverse time, which represents the reverse time range to be counted for the current time and is set to 60 minutes;

[0037] M0 represents the weight value of the current ratio, and the default value is 1. When M0 is larger, the weight of the current ratio is larger, and the statistics will quickly approach the current ratio.

[0038] 120 means the calculation cycle is 500ms, and the calculation is performed 120 times in 1 minute;

[0039] Y0 represents the memory, statistics and calculation of the average ratio of the previous period.

[0040] Beneficial effects:

[0041] 1. The present invention adopts a pure software programming method, and writes a logic algorithm in the gas flow control module of the original DCS system to achieve it. It does not require any additional equipment investment, has low cost and good effect.

[0042] 2. Adopting automatic calculation based on genetic algorithm and big data statistics, the dead zone of gas regulating valve can be adaptively overcome, especially the reverse dead zone of gas regulating valve, so that the adjustment is more timely and the problems of too many switches, slow callback and lag in the prior art can be avoided.

[0043] 3. Compared with the traditional PID control method, the adjustment lag is smaller, the control accuracy is higher, and the system robustness is better.

[0044] 4. The present invention has been successfully applied and implemented in the combustion of multiple gas-controlled boilers, with good optimization control effect, outstanding economic benefits and promotion value. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 is a flow chart of the method of the present invention;

[0046] Figure 2 It is a flow chart of the gas opening statistics learning program of the present invention;

[0047] Figure 3 This is a flow chart of the statistical learning program of the gas flow and load ratio in the present invention. DETAILED DESCRIPTION

[0048] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the present invention will be briefly introduced below in combination with the drawings and the description of the embodiments or the prior art. Obviously, the following description of the structure of the drawings is only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. It should be noted that the description of these embodiments is used to help understand the present invention, but does not constitute a limitation of the present invention.

[0049] Example:

[0050] like Figure 1 As shown, this embodiment provides a method for overcoming the reverse regulation dead zone of a gas regulating valve by using big data statistics, including the following steps:

[0051] Predict the direction of valve movement;

[0052] According to the expected movement direction of the valve, the valve adjustment command is forced to switch to the average opening of the previous 10-minute time range and wait for new valve adjustment commands.

[0053] Using big data statistical methods, the average opening instruction of the valve within 10 minutes is calculated in real time;

[0054] According to the prediction of the valve action direction, the valve adjustment command is limited to the average opening command to wait for the new valve PID adjustment command, and the valve is allowed to wait for 0.5 seconds at the new opening position to judge the new valve adjustment command; avoid over-adjustment, excessive switching and oscillation, overcome the problems of slow callback and lag, and thus overcome the reverse adjustment dead zone of the valve. Using big data statistical methods, the average valve opening command of the valve opening in the reverse time (10 minutes) is calculated in real time. When the load and gas flow change direction are predicted to be reversed, the valve opening command is directly preset and tracked to the average opening command calculated previously, and a reverse dead zone is superimposed, so that the valve is allowed to wait for new commands at the new position, thereby reducing and effectively overcoming the reverse adjustment dead zone of the valve. The valve opening command is consistent with the valve opening feedback when it is normal, and the deviation is very small. It is not normal if it is greater than 1% (adjustment dead zone) and needs to be adjusted to be consistent. The valve is operated according to the opening command. The average opening feedback or average opening command is a concept. The command does not have an adjustment dead zone, and the feedback must overcome the dead zone and can only act, and the feedback will change.

[0055] Using big data statistical methods, the average ratio of gas flow to load within 60 minutes is calculated in real time;

[0056] After the load target is determined, the gas flow required for the current load target is obtained by multiplying the average ratio of the statistical gas flow to the load by the load target, and then the gas flow target required for the current load can be obtained;

[0057] The expected valve opening can be obtained based on the gas flow target and the flow characteristics of the valve;

[0058] According to the reverse regulation dead zone of the gas regulating valve, the opening range of the gas valve can be determined, and the valve command can be directly limited to this range, thereby avoiding excessive switching of the regulating valve and over-regulation of the gas flow. The regulation dead zone is the dead zone presented on site according to the actual switch. Under normal circumstances, after the valve is switched, the controlled gas flow will increase or decrease. If the valve is switched, the gas flow does not change, which is the regulation dead zone. If the valve is opened and then closed, or closed and then opened, it is the reverse regulation dead zone. This is a characteristic that is difficult to overcome for gas valves with large pipeline diameters!

[0059] Utilizing big data statistical methods, the average relationship between load and gas flow (coal consumption) in one hour of reverse time is calculated in real time. After the load target is determined, multiplying it by the statistical gas flow is the gas flow target required by the current load. The valve opening can be known based on the gas flow target and the flow characteristics of the valve. Based on the reverse adjustment dead zone of the gas regulating valve, the opening range of the gas valve can be determined (the upper and lower limits of the opening, which change dynamically with the load and coal consumption), thereby avoiding excessive switching of the regulating valve and over-adjustment of the gas flow.

[0060] In another embodiment of the present invention, a method for predicting a valve movement direction comprises the following steps:

[0061] Adjust the valve opening command according to the deviation between the actual load of the unit and the load target;

[0062] The valve movement direction is determined based on the average weighted change rate of the opening command in the past 3-5s. If the average weighted change rate is greater than 0, the opening command becomes larger; if the average weighted change rate is less than 0, the opening command becomes smaller.

[0063] In another embodiment of the present invention, the big data statistical method includes an average opening instruction statistical method and an average ratio statistical method of gas flow and load.

[0064] In another embodiment of the present invention, Figure 2 As shown, the average opening instruction statistical method includes the following steps:

[0065] Determine whether the signal quality of the gas flow I1 is normal. When the signal quality is abnormal, it does not participate in the gas flow I1 data statistics; when the signal quality is normal, perform data statistics and learning on the gas valve opening I2 to obtain the statistical initial value Y1; the gas flow has normal upper and lower limits and a normal fluctuation range. By comparing the gas flow with its normal upper and lower limits, if it exceeds the limit, it means that the signal is abnormal and the signal quality is bad. If it does not exceed the limit, it is normal. Similarly, determine the fluctuation value of the gas signal in each cycle and compare it with the normal fluctuation upper limit. If it exceeds the limit, it also means that the signal is abnormal and the signal quality is bad. Each valve has its flow characteristics, and the opening is proportional to the flow, which is obtained according to the actual flow characteristics of the gas valve on site, or according to the valve characteristic test.

[0066] Calculate and obtain the average opening O1;

[0067] Statistics and calculation of the average opening Y of the previous cycle, Y = O1;

[0068] After this cycle statistics are completed, the next cycle statistics will be repeated.

[0069] In another embodiment of the present invention, the average opening O1 = (RT*120*Y+M*Y1) / (RT*120+M);

[0070] RT stands for reverse time, which represents the reverse time range to be counted for the current time and is set to 10 minutes;

[0071] M represents the weight of the current ratio, and the default value is 1. When M is larger, the weight of the current ratio is larger, and the statistics will quickly approach the current ratio.

[0072] 120 means the calculation cycle is 500ms, and the calculation is performed 120 times in 1 minute;

[0073] Y represents the memorized, statistically calculated average opening value of the previous period.

[0074] In another embodiment of the present invention, Figure 3 As shown, the statistical method of the average ratio of gas flow to load includes the following steps:

[0075] Determine whether the signal quality of the gas flow I1 and the load I3 is normal. If the signal quality is abnormal, the data statistics of the gas flow I1 and the load I3 are not included. If the signal quality is normal, the data statistics and learning of the gas flow I1 and the load I3 are performed to obtain the initial statistical value Y2, that is, the normal valve opening value I1 / I3;

[0076] The average ratio O2 was calculated;

[0077] Statistics and calculation of the average ratio Y0 of the previous period, Y0 = O2;

[0078] After this cycle statistics are completed, the next cycle statistics will be repeated.

[0079] In another embodiment of the present invention, the average ratio O2=(RT*120*Y0+M0*Y2) / (RT*120+M0);

[0080] RT stands for reverse time, which represents the reverse time range to be counted for the current time and is set to 60 minutes;

[0081] M0 represents the weight value of the current ratio, and the default value is 1. When M0 is larger, the weight of the current ratio is larger, and the statistics will quickly approach the current ratio.

[0082] 120 means the calculation cycle is 500ms, and the calculation is performed 120 times in 1 minute;

[0083] Y0 represents the memory, statistics and calculation of the average ratio of the previous period.

[0084] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the scope of protection of the present invention.

Claims

1. A method for overcoming the reverse regulation dead zone of a gas regulating valve by using big data statistics, characterized in that: The following steps are involved: Predict the direction of valve movement; According to the expected action direction of the valve, the valve adjustment command is forced to switch to the average opening of the past 10 minutes and wait for new valve adjustment commands; Using big data statistical methods, the average opening instruction of the valve within 10 minutes is calculated in real time; According to the prediction of valve movement direction, the valve adjustment command is limited to the average opening command, and the new valve PID adjustment command is waited again; Using big data statistical methods, the average ratio of gas flow to load within 60 minutes is calculated in real time; After the load target is determined, the gas flow required for the current load target is obtained by multiplying the average ratio of the statistical gas flow to the load by the load target, and then the gas flow target required for the current load can be obtained; The expected valve opening can be obtained based on the gas flow target and the flow characteristics of the valve; The opening range of the gas valve can be determined based on the reverse adjustment dead zone of the gas regulating valve.

2. The method for overcoming the reverse regulation dead zone of the gas regulating valve by using big data statistics according to claim 1 is characterized in that: The method for predicting the valve movement direction comprises the following steps: According to the deviation between the actual load of the unit and the load target, adjust the opening command of the gas regulating valve; The valve movement direction is determined based on the average weighted change rate of the opening command in the past 3-5s. If the average weighted change rate is greater than 0, the opening command becomes larger; if the average weighted change rate is less than 0, the opening command becomes smaller.

3. The method for overcoming the reverse regulation dead zone of the gas regulating valve by using big data statistics according to claim 1 is characterized in that: The big data statistical method includes an average opening instruction statistical method and an average ratio statistical method of gas flow and load.

4. The method for overcoming the reverse regulation dead zone of the gas regulating valve by using big data statistics according to claim 3 is characterized in that: The average opening instruction statistical method comprises the following steps: Determine whether the signal quality of the gas flow I1 is normal. If the signal quality is abnormal, no average valve position statistics are performed. If the signal quality is normal, data statistics and learning of the gas valve opening I2 are performed to obtain the initial statistical value Y1. Calculate and obtain the average opening O1; Statistics and calculation of the average opening Y of the previous cycle, Y = O1; After this cycle statistics are completed, the next cycle statistics will be repeated.

5. The method for overcoming the reverse regulation dead zone of the gas regulating valve by using big data statistics according to claim 4 is characterized in that: The average opening O1 = (RT*120*Y+M*Y1) / (RT*120+M); RT stands for reverse time, which represents the reverse time range to be counted for the current time and is set to 10 minutes; M represents the weight of the current ratio, and the default value is 1. When M is larger, the weight of the current ratio is larger, and the statistics will quickly approach the current ratio. 120 means the calculation cycle is 500ms, and the calculation is performed 120 times in 1 minute; Y represents the memorized, statistically calculated average opening value of the previous period.

6. The method for overcoming the reverse regulation dead zone of the gas regulating valve by using big data statistics according to claim 3 is characterized in that: The method for calculating the average ratio of gas flow to load comprises the following steps: Determine whether the signal quality of the gas flow I1 and the load I3 is normal. When the signal quality is abnormal, the data statistics of the gas flow I1 and the load I3 are not included; when the signal quality is normal, the ratio data of the gas flow I1 and the load I3 is counted and learned to obtain the initial statistical value Y2, that is, the normal valve opening value I1 / I3; The average ratio O2 was calculated; Statistics and calculation of the average ratio Y0 of the previous period, Y0 = O2; After this cycle statistics are completed, the next cycle statistics will be repeated.

7. The method for overcoming the reverse regulation dead zone of the gas regulating valve by using big data statistics according to claim 6 is characterized in that: The average ratio O2=(RT*120*Y0+M0*Y2) / (RT*120+M0); RT stands for reverse time, which represents the reverse time range to be counted for the current time and is set to 60 minutes; M0 represents the weight value of the current ratio, and the default value is 1. When M0 is larger, the weight of the current ratio is larger, and the statistics will quickly approach the current ratio. 120 means the calculation cycle is 500ms, and the calculation is performed 120 times in 1 minute; Y0 represents the memory, statistics and calculation of the average ratio of the previous period.

Images