On-line monitoring and inhibiting system applied to cold convertor station valve hall air conditioner waterway corrosion

Abstract

The invention discloses an on-line monitoring and inhibiting system applied to cold convertor station valve hall air conditioner waterway corrosion, and relates to the technical field of power transformation on-line monitoring and overhauling. The system comprises an ethylene glycol concentration management module, a corrosion inhibitor management module, an ion concentration management module and an operation and maintenance data analysis module; the operation and maintenance data analysis module is used for judging whether ethylene glycol needs to be added or not according to the concentration of the ethylene glycol, calculating to obtain a first target quantity when the first judgment result is that the ethylene glycol needs to be added, judging whether a corrosion inhibitor needs to be added or not according to the PH value of the solution, and calculating to obtain a second target quantity when the second judgment result is that the corrosion inhibitor needs to be added; the controller is also used for judging a corrosion judgment result of the corrosion degree according to the iron ion content and the oxygen content of the solution; in this way, operation and maintenance personnel can be helped to find the corrosion state of the valve hall air-conditioning waterway system of the cold convertor station in time, automatic adding of the anti-freezing solution and the slow-release agent is achieved, and time and labor are saved in monitoring and restraining of corrosion of the valve hall air-conditioning waterway system.

Description

Technical Field

[0001] The present invention relates to the technical field of substation online monitoring and maintenance, and more specifically, to an online monitoring and suppression system for corrosion of air-conditioning water channels in valve halls of cold converter stations. Background Art

[0002] The converter valve is the core equipment in the converter station. Each valve hall in the converter station is equipped with two sets of valve hall air conditioning units, which are used to keep the converter valve in the valve hall within the appropriate temperature range and work stably for a long time. In cold areas (winter temperature is below zero degrees Celsius), the converter station valve hall air conditioning will add a certain amount of ethylene glycol solution to its cold water circuit to prevent ice from freezing on the water circuit and damaging the equipment in order to ensure its normal operation in winter. At present, ethylene glycol solution coolant is widely used in the valve hall air conditioning system of converter stations in cold areas. However, in actual applications, it is found that ethylene glycol has strong corrosiveness to carbon steel, cast iron, solder in the air conditioning water circuit, and even the copper tube of the valve hall air conditioning screw heat exchanger. There have been more than one cases where the valve hall air conditioning water pipeline or the air conditioning screw heat exchanger copper tube has been severely corroded, resulting in the valve hall air conditioning unable to cool the converter valve normally. If it is not discovered in time, the temperature in the converter valve will be too high, causing the valve element to fail. In severe cases, the valve group will be directly locked, causing the DC system to lose power, causing interference to the entire system, and affecting the stable operation of the entire DC system.

[0003] In order to solve the problem of ethylene glycol solution corrosive to valve cooling equipment, the industry usually adds a certain amount of corrosion inhibitor to alleviate its corrosion to the system; however, since both ethylene glycol solution and corrosion inhibitor will be consumed during the continuous circulation of the water channel, new ethylene glycol and corrosion inhibitor need to be added at intervals to ensure the antifreeze effect.

[0004] The current method of adding is manual addition, which requires operation and maintenance personnel to go to the site and repeatedly perform the following steps: first, use a refractometer to measure the ethylene glycol concentration after sampling, then add a certain amount of ethylene glycol to the system and wait for the circulation pump to evenly diffuse the ethylene glycol throughout the system, then take another sample and use a refractometer to measure the ethylene glycol concentration. If the concentration does not meet the standard, continue the above steps until it meets the requirements. The above process is very time-consuming and labor-intensive, and the on-site manual control process is often inaccurate, resulting in ethylene glycol concentration exceeding or insufficient; adding ethylene glycol solution corrosion inhibitors also has the same problem. Summary of the invention

[0005] The object of the present invention is to provide an online monitoring and suppression system for corrosion of air-conditioning water channels in valve halls of cold converter stations, so as to solve the problems existing in the above-mentioned background technology.

[0006] The above technical objectives of the present invention are achieved through the following technical solutions: In the first aspect, the present application provides an online monitoring and suppression system for corrosion of air conditioning water channels in valve halls of cold converter stations, comprising: The ethylene glycol concentration management module includes an ethylene glycol concentration monitoring unit and an ethylene glycol filling unit, wherein: the ethylene glycol concentration monitoring unit is used to obtain the concentration of ethylene glycol in the air conditioning water circuit; the ethylene glycol filling unit is used to add a first target amount of ethylene glycol to the air conditioning water circuit; The corrosion inhibitor management module includes a pH value monitoring unit and a corrosion inhibitor filling unit, wherein: the pH value monitoring unit is used to obtain the pH value of the solution in the air-conditioning water channel; the corrosion inhibitor filling unit is used to add a second target amount of corrosion inhibitor to the air-conditioning water channel; The ion concentration management module includes an iron ion concentration monitoring unit and an oxygen content monitoring unit, wherein: the iron ion concentration monitoring unit is used to obtain the iron ion content in the air conditioning water path, and the oxygen content monitoring unit is used to obtain the oxygen content of the solution in the air conditioning water path; The operation and maintenance data analysis module is used to determine whether ethylene glycol needs to be added according to the concentration of ethylene glycol, and calculate the first target amount when the first judgment result is that addition is required, and to determine whether corrosion inhibitor needs to be added according to the pH value of the solution, and calculate the second target amount when the second judgment result is that addition is required; it is also used to determine the corrosion judgment result of the degree of corrosion of the air-conditioning water channel according to the iron ion content and the oxygen content of the solution.

[0007] The beneficial effects of the present invention are as follows: in this scheme, firstly, the concentration of ethylene glycol in the air-conditioning water circuit is obtained in real time through the ethylene glycol concentration monitoring unit. The judgment process of ethylene glycol is to compare the real-time monitored ethylene glycol concentration with the ethylene glycol standard concentration. If the ethylene glycol concentration is less than the ethylene glycol standard concentration, it indicates that ethylene glycol needs to be added to the water circuit. Similarly, the judgment process of the corrosion inhibitor is the same; secondly, when ethylene glycol needs to be added or corrosion inhibitor needs to be added, the operation and maintenance data analysis module is used to calculate the first target amount of ethylene glycol to be added or the second target amount of corrosion inhibitor to be added, so that ethylene glycol can achieve a good antifreeze effect and the corrosion inhibitor can achieve a good corrosion inhibition effect; finally, the degree of corrosion of the equipment components in the air-conditioning water circuit is judged by the real-time obtained iron ion content and the oxygen content of the solution in the air-conditioning water circuit. Since the iron ion content and the oxygen content in the air-conditioning water circuit can reflect the degree of corrosion of the equipment components, the judgment process can set an early warning range. When these two values ​​enter this range, it indicates that the corrosion is serious and maintenance work needs to be carried out.

[0008] In this scheme, the fundamental purpose is to monitor the corrosion status of the air-conditioning water system in the valve hall of the cold converter station in real time and to automatically inhibit corrosion in time; first, when the concentration of the ethylene glycol solution in the valve hall water system does not reach the set standard value, the ethylene glycol concentration management module of the system will detect and automatically adjust the ethylene glycol concentration in the water through the ethylene glycol filling unit until it reaches the set value; second, as time goes by, the corrosion inhibitor concentration in the water system will decrease with the consumption concentration, and the pH value of the solution will also decrease. In this way, it is impossible to keep the entire water solution in a weakly alkaline environment to inhibit corrosion. At this time, the corrosion inhibitor management module of the system will detect and control the corrosion inhibitor filling unit. To supplement the corrosion inhibitor, so that the pH value of the solution is maintained near the standard value. Even if the entire aqueous solution is weakly alkaline, it can effectively inhibit the aggravation of the corrosion effect caused by ethylene glycol and realize the corrosion inhibition function; thirdly, while realizing the above functions, the system can also provide early warning and alarm for the dissolved oxygen concentration and trivalent iron ion concentration that reflect the degree of corrosion of the air-conditioning water system, so as to facilitate the operation and maintenance personnel to learn about the corrosion of the valve hall air-conditioning water system; through the above series of means, it can help the operation and maintenance personnel to timely discover the corrosion status of the valve hall air-conditioning water system in the cold converter station, and realize the automatic addition of antifreeze and slow-release agent, making the monitoring and inhibition of the corrosion of the valve hall air-conditioning water system time-saving and labor-saving.

[0009] Based on the above technical solution, the present invention can also be improved as follows.

[0010] Furthermore, the operation and maintenance data analysis module includes a target quantity calculation unit, which is used to calculate the first target quantity and the second target quantity, specifically: When the first judgment result indicates that ethylene glycol needs to be added, a first difference is calculated based on the concentration of ethylene glycol corresponding to the first judgment result and a preset ethylene glycol standard concentration, and the first difference is input into the ethylene glycol prediction model for processing to obtain a first target amount; When the second judgment result is that corrosion inhibitor needs to be added, the pH value of the solution corresponding to the second judgment result is calculated with the preset standard pH value to obtain a second difference, and the second difference is input into the corrosion inhibitor prediction model for processing to obtain a second target amount.

[0011] The beneficial effect of adopting the above further scheme is that the corrosion inhibitor prediction model can be a currently common neural network model, which has a strong feature learning ability, so as to learn the characteristic relationship between the increase in the amount of ethylene glycol or corrosion inhibitor and the change in ethylene glycol concentration or pH value in the air-conditioning water circuit, and finally the calculated concentration difference or the calculated pH value difference can be used to calculate the amount that needs to be increased. This method can accurately calculate the amount of ethylene glycol or corrosion inhibitor that needs to be added, avoiding the current situation where operation and maintenance personnel need to add it repeatedly.

[0012] Furthermore, the above-mentioned ethylene glycol prediction model or corrosion inhibitor prediction model is obtained by the following method: Obtain the corresponding addition amount, current value, and actual value after addition for each time in the historical addition record. The historical addition record is the ethylene glycol addition record or the corrosion inhibitor addition record; The change value is calculated based on the corresponding current value each time and the actual value after addition, and the preset initial model is trained using the corresponding change value and addition amount each time until the preset initial model reaches the training end condition, and the preset initial model that reaches the training end condition is determined as the ethylene glycol prediction model or the corrosion inhibitor prediction model.

[0013] The beneficial effect of adopting the above further scheme is that the ethylene glycol prediction model and the corrosion inhibitor prediction model are obtained through the same training method, and the ethylene glycol prediction model can learn the characteristic relationship between the increase in ethylene glycol and the change in ethylene glycol concentration in the air-conditioning water circuit, and the corrosion inhibitor prediction model can learn the characteristic relationship between the increase in corrosion inhibitor and the change in pH value in the air-conditioning water circuit.

[0014] Furthermore, during the training process of the above preset initial model, each iteration is specifically as follows: After the change value is input into the preset initial model, the output of each neuron in each layer of the network in the preset initial model is calculated; Based on the output of each neuron and the preset adjustable parameters, the gradient of the output of each neuron is calculated, and based on the gradient of the output of each neuron and the preset adjustable parameters, the gradient of the connection strength between each neuron is calculated; Based on the gradient of the connection strength between each neuron, the connection strength between each neuron is updated, and after the connection strength between each neuron is updated, the preset adjustable parameters are readjusted.

[0015] The beneficial effect of adopting the above further scheme is that the neural network in the preset initial model has a multi-layer structure, and each iteration process is to adjust the various model parameters of the model, including connection strength, etc. after the change value is input into the preset initial model. This can ensure that more neurons participate in learning in the feature learning device, so as to achieve higher accuracy after learning.

[0016] Furthermore, the output of each neuron in each layer of the network in the above preset initial model is specifically: ; In the formula, Indicates t The iteration l Neurons in a layer network i The output, Indicates l Neurons in a layer networki With neurons j The connection strength between represents the activation function, and the neuron i After output ， ,otherwise ; represents the parameter that controls the change of neuronal membrane potential over time, Indicates t This iteration l Neurons in a layer network i The membrane potential before output, Indicates t This iteration l Neurons in a layer network i Input, represents the upper threshold of membrane potential, Represents the Heaviside function.

[0017] Furthermore, the gradient of the output of each of the above neurons is specifically: ,in: , ; In the formula, Indicates l Neurons in a layer network i The gradient of the output of Indicates l- Neurons in a 1-layer network i The gradient of the output of Indicates l Neurons in a layer network i The gradient of the input, represents the loss function of connection strength, , A represents the intermediate quantity along the spatial direction, B represents the intermediate quantity along the time direction, All represent preset adjustable parameters. Indicates t This iteration l Neurons in a layer network i The membrane potential before output, Indicates t This iteration l Neurons in a layer network i The membrane potential after output, represents the upper threshold of membrane potential, Represents the parameter that controls the temporal variation of the neuron's membrane potential.

[0018] Furthermore, the gradient of the connection strength between the above neurons is specifically: ; In the formula, Indicates l Neurons in a layer network i With neurons j The gradient of the connection strength between Indicates l Neurons in a layer network i The gradient of the output of All represent preset adjustable parameters. Indicates t This iteration l Neurons in a layer network i Input, Indicates t This iteration l Neurons in a layer network i The membrane potential after output, Represents the upper threshold of membrane potential.

[0019] Furthermore, the connection strength between each neuron is updated as follows: ; In the formula, No. l Neurons in a layer network i With neurons j The change in the connection strength between them after the update, represents the learning rate of the model, Indicates l Neurons in a layer network i With neurons j The gradient of the connection strength between A loss function representing the strength of the connection.

[0020] Furthermore, the above preset adjustable parameters are specifically: , ; In the formula, All represent preset adjustable parameters. Indicates preset adjustable parameters The minimum value of Indicates preset adjustable parameters The maximum value of T represents the maximum number of iterations, t Indicates the current iteration number.

[0021] In a second aspect, the present application provides the application of an online monitoring and suppression system for corrosion of air-conditioning water circuits in valve halls of cold converter stations as described in any one of the first aspects in air-conditioning water circuits.

[0022] Compared with the prior art, the present invention has at least the following beneficial effects: In the present application, firstly, the concentration of ethylene glycol in the air-conditioning water circuit is obtained in real time through the ethylene glycol concentration monitoring unit. The judgment process of ethylene glycol is to compare the real-time monitored ethylene glycol concentration with the standard ethylene glycol concentration. If the ethylene glycol concentration is less than the standard ethylene glycol concentration, it indicates that ethylene glycol needs to be added to the water circuit. Similarly, the judgment process of the corrosion inhibitor is the same; secondly, when ethylene glycol needs to be added or corrosion inhibitor needs to be added, the operation and maintenance data analysis module is used to calculate the first target amount of ethylene glycol to be added or the second target amount of corrosion inhibitor to be added, so that ethylene glycol can achieve a good antifreeze effect and the corrosion inhibitor can achieve a good corrosion inhibition effect; finally, the degree of corrosion of the equipment and components in the air-conditioning water circuit is judged by the real-time obtained iron ion content and the oxygen content of the solution in the air-conditioning water circuit. Since the iron ion content and oxygen content in the air-conditioning water circuit can reflect the degree of corrosion of the equipment and components, the judgment process can set an early warning range. When these two values ​​enter this range, it indicates that the corrosion is serious and maintenance work needs to be carried out.

[0023] In the present application, the fundamental purpose is to monitor the corrosion status of the air-conditioning water system in the valve hall of the cold converter station in real time and to automatically inhibit corrosion in time; first, when the concentration of the ethylene glycol solution in the valve hall water system does not reach the set standard value, the ethylene glycol concentration management module of the system will discover and automatically adjust the ethylene glycol concentration in the water through the ethylene glycol filling unit until it reaches the set value; second, as time goes by, the corrosion inhibitor concentration in the water system will decrease with the consumption concentration, and the pH value of the solution will also decrease accordingly, so that the entire water solution cannot maintain a weakly alkaline environment to inhibit corrosion. At this time, the corrosion inhibitor management module of the system will discover and control the corrosion inhibitor filling unit. To supplement the corrosion inhibitor, so that the pH value of the solution is maintained near the standard value. Even if the entire aqueous solution is weakly alkaline, it can effectively inhibit the aggravation of the corrosion effect caused by ethylene glycol and realize the corrosion inhibition function; thirdly, while realizing the above functions, the system can also provide early warning and alarm for the dissolved oxygen concentration and trivalent iron ion concentration that reflect the degree of corrosion of the air-conditioning water system, so as to facilitate the operation and maintenance personnel to learn about the corrosion of the valve hall air-conditioning water system; through the above series of means, it can help the operation and maintenance personnel to timely discover the corrosion status of the valve hall air-conditioning water system in the cold converter station, and realize the automatic addition of antifreeze and slow-release agent, making the monitoring and inhibition of the corrosion of the valve hall air-conditioning water system time-saving and labor-saving.

[0024] In the present application, the corrosion inhibitor prediction model can be a currently common neural network model, which has a strong feature learning ability, so as to learn the characteristic relationship between the increase in the amount of ethylene glycol or corrosion inhibitor and the change in the concentration of ethylene glycol or the change in pH value in the air-conditioning water circuit, and finally the amount to be increased can be calculated using the calculated concentration difference or the calculated pH value difference. This method can accurately calculate the amount of ethylene glycol or corrosion inhibitor that needs to be added, avoiding the current situation where the operation and maintenance personnel need to add it repeatedly. The ethylene glycol prediction model and the corrosion inhibitor prediction model are obtained by the same training method, and the ethylene glycol prediction model can learn the characteristic relationship between the increase in the amount of ethylene glycol and the change in the concentration of ethylene glycol in the air-conditioning water circuit, and the corrosion inhibitor prediction model can learn the characteristic relationship between the increase in the amount of corrosion inhibitor and the change in pH value in the air-conditioning water circuit. The neural network in the preset initial model has a multi-layer structure. Each iteration process is to adjust the various model parameters of the model, including the connection strength, after the change value is input into the preset initial model. This can ensure that more neurons participate in learning in the feature learning device, so as to achieve higher accuracy after learning. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] The drawings described herein are used to provide a further understanding of the embodiments of the present invention, constitute a part of this application, and do not constitute a limitation of the embodiments of the present invention. In the drawings: Figure 1 It is a connection diagram of the online monitoring and simulation system in the embodiment of the present invention. DETAILED DESCRIPTION

[0026] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0027] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0028] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, further definition and explanation thereof is not required in subsequent drawings.

[0029] In the description of the embodiments of the present invention, "plurality" means at least 2.

[0030] Embodiment 1: In order to solve the problem that the current addition method is manual addition, and the addition process is time-consuming and labor-intensive, and the on-site manual control process is often inaccurate, resulting in excessive or insufficient ethylene glycol concentration or pH value, this embodiment provides an online monitoring and suppression system for corrosion of air conditioning waterways in valve halls of cold converter stations, such as Figure 1 As shown, including: The ethylene glycol concentration management module includes an ethylene glycol concentration monitoring unit and an ethylene glycol filling unit, wherein: the ethylene glycol concentration monitoring unit is used to obtain the concentration of ethylene glycol in the air-conditioning water circuit; the ethylene glycol filling unit is used to add a first target amount of ethylene glycol to the air-conditioning water circuit.

[0031] Among them, the ethylene glycol concentration monitoring unit can be installed at the head end, middle end and end of the valve hall air-conditioning water channel through a connector, and an ethylene glycol concentration real-time monitoring sensor is responsible for providing real-time ethylene glycol concentration information of the water channel system, and storing the real-time data records as historical data that can be queried; specifically, the real-time concentration information output by the ethylene glycol concentration monitoring unit is the middle of three, for example, assuming that the concentrations uploaded by the three sensors are X51>X52>X53, then the real-time concentration output by the ethylene glycol concentration monitoring unit is X52; it can also be when one of X51, X52, and X53 is greater than a preset value When W, the system shields sensor X51 or X53 and then reports an alarm to the superior. The real-time concentration value is the average value of the remaining two sensor concentrations and uploaded. If two of X51, X52, and X53 are greater than W, the three values ​​X51, X52, and X53 are compared with the historical ethylene glycol data of the previous point, and the value with the smallest error in the uploaded record is uploaded; this prevents the concentration value of ethylene glycol from mutating. Therefore, each uploaded record of ethylene glycol concentration value must be compared with the ethylene glycol data of other points uploaded within 24 hours. If the difference exceeds Z%, an alarm is issued to the superior for maintenance and other operations, that is, it can be achieved through Figure 1 The evaluation alarm module is implemented in .

[0032] Among them, next to the real-time monitoring sensors of ethylene glycol concentration at the head, middle and end of the air-conditioning water channel in the valve hall, refractometers can be installed to enable the operation and maintenance personnel in the station to check the three ethylene glycol concentrations on-site, so that the operation and maintenance personnel can check the ethylene glycol concentrations at the head, middle and end of the system on-site, and compare them with the ethylene glycol concentration in the ethylene glycol concentration monitoring unit.

[0033] Furthermore, the above system may also include a corrosion inhibitor management module, which includes a pH value monitoring unit and a corrosion inhibitor filling unit, wherein: the pH value monitoring unit is used to obtain the pH value of the solution in the air-conditioning water circuit; the corrosion inhibitor filling unit is used to add a second target amount of corrosion inhibitor to the air-conditioning water circuit.

[0034] It should be noted that, for the ethylene glycol replenisher and the corrosion inhibitor replenisher, the concentrations in the solutions are consistent in this embodiment, in order to facilitate the training of the model below. When the concentration of the replenisher changes, the model below needs to be readjusted; and for the addition of the ethylene glycol replenisher and the corrosion inhibitor replenisher, it can be achieved through a pump and a liquid flow meter, that is, the flow meter and the pump are connected to each other, and the output end of the pump is connected to the air-conditioning water circuit. When the amount to be added is obtained, the flow meter controls the output of the corresponding amount, and then enters the air-conditioning water circuit through the pump.

[0035] Among them, the pH value monitoring unit can be installed at the head end, middle end and end of the valve hall air-conditioning water channel through a joint. A pH value real-time monitoring sensor is responsible for providing pH value information in the water channel and recording and storing real-time data as historical data that can be queried; specifically, the real-time concentration information output by the pH value monitoring unit is the middle of three, for example, if the concentrations uploaded by the three sensors are A51>A52>A53, then the pH value output by the pH value monitoring unit is A52; it can also be when one of A51, A52, A53 is greater than the preset value S At 1, the system shields sensor A51 or A53 and then reports an alarm to the superior. The pH value is the average value of the remaining two sensors and uploaded. If two of A51, A52, and A53 are greater than S1, the three values ​​A51, A52, and A53 are compared with the historical pH value data of the previous point, and the value with the smallest error in the uploaded record is uploaded; this will prevent the pH value from mutating. Therefore, the pH value of each uploaded record is compared with the pH value of other points recorded within 24 hours. If the difference exceeds S2%, an alarm is issued to the superior for maintenance and other operations, that is, through Figure 1 The evaluation alarm module is implemented in .

[0036] Furthermore, the above system may also include an ion concentration management module, which includes an iron ion concentration monitoring unit and an oxygen content monitoring unit, wherein: the iron ion concentration monitoring unit is used to obtain the iron ion content in the air-conditioning water circuit, and the oxygen content monitoring unit is used to obtain the oxygen content of the solution in the air-conditioning water circuit.

[0037] Among them, the data uploading method of the iron ion concentration monitoring unit and the oxygen content monitoring unit can also adopt the above-mentioned ethylene glycol and pH value data uploading method, that is, obtaining and judging through three sensors, which will not be repeated here; because the iron ion content and oxygen content in the air-conditioning water circuit can reflect the degree of corrosion of the equipment components, the judgment process can set an early warning range. When these two values ​​enter this range, an alarm is issued, indicating that the corrosion is serious and maintenance and other work are required.

[0038] Furthermore, the above system may also include an operation and maintenance data analysis module, which is used to determine whether ethylene glycol needs to be added according to the concentration of ethylene glycol, and calculate a first target amount when the first judgment result is that addition is required, and to determine whether a corrosion inhibitor needs to be added according to the pH value of the solution, and calculate a second target amount when the second judgment result is that addition is required; and is also used to determine a corrosion judgment result of the degree of corrosion of the air-conditioning water channel according to the iron ion content and the oxygen content of the solution.

[0039] Optionally, the operation and maintenance data analysis module includes a target quantity calculation unit, which is used to calculate the first target quantity and the second target quantity, specifically: S1, when the first judgment result is that ethylene glycol needs to be added, a first difference is calculated based on the concentration of ethylene glycol corresponding to the first judgment result and the preset ethylene glycol standard concentration, and the first difference is input into the ethylene glycol prediction model for processing to obtain a first target amount.

[0040] S2, when the second judgment result is that corrosion inhibitor needs to be added, the pH value of the solution corresponding to the second judgment result is calculated with the preset standard pH value to obtain a second difference, and the second difference is input into the corrosion inhibitor prediction model for processing to obtain a second target amount.

[0041] Among them, the corrosion inhibitor prediction model can be a common neural network model currently, such as a deep pulse neural network, which has a strong feature learning ability, so as to learn the characteristic relationship between the increase in the amount of ethylene glycol or corrosion inhibitor and the change in the ethylene glycol concentration or the pH value in the air-conditioning water circuit, and finally the calculated concentration difference or the calculated pH value difference can be used to calculate the amount that needs to be increased. This method can accurately calculate the amount of ethylene glycol or corrosion inhibitor that needs to be added, avoiding the current situation where operation and maintenance personnel need to add it repeatedly.

[0042] Optionally, the above ethylene glycol prediction model or corrosion inhibitor prediction model is obtained by the following method: S31, obtaining the corresponding addition amount, current value, and actual value after addition in each historical addition record, where the historical addition record is an ethylene glycol addition record or a corrosion inhibitor addition record.

[0043] S32, calculating the change value according to the current value corresponding to each time and the actual value after addition, and training the preset initial model using the change value corresponding to each time and the addition amount until the preset initial model reaches the training end condition, and determining the preset initial model that reaches the training end condition as the ethylene glycol prediction model or the corrosion inhibitor prediction model.

[0044] Among them, the ethylene glycol prediction model and the corrosion inhibitor prediction model are obtained through the same training method, and the ethylene glycol prediction model can learn the characteristic relationship between the increase in ethylene glycol and the change in ethylene glycol concentration in the air-conditioning water circuit, and the corrosion inhibitor prediction model can learn the characteristic relationship between the increase in corrosion inhibitor and the change in pH value in the air-conditioning water circuit.

[0045] Optionally, during the training process of the preset initial model, each iteration is specifically as follows: S321, after the change value is input into the preset initial model, the output of each neuron in each layer of the network in the preset initial model is calculated.

[0046] Optionally, the output of each neuron in each layer of the network in the above preset initial model is specifically: ; In the formula, Indicates t The first iteration l Neurons in a layer network i The output, Indicates l Neurons in a layer network i With neurons j The connection strength between represents the activation function, and the neuron i After output ， ,otherwise ; represents the parameter that controls the change of neuronal membrane potential over time, Indicates t This iteration l Neurons in a layer network i The membrane potential before output, Indicates t This iteration l Neurons in a layer network i Input, represents the upper threshold of membrane potential, Represents the Heaviside function.

[0047] S322, based on the output of each neuron and the preset adjustable parameters, the gradient of the output of each neuron is calculated, and according to the gradient of the output of each neuron and the preset adjustable parameters, the gradient of the connection strength between each neuron is calculated.

[0048] Optionally, the gradient of the output of each of the above neurons is specifically: ,in: , ; In the formula, Indicates l Neurons in a layer network i The gradient of the output of Indicates l- Neurons in a 1-layer network i The gradient of the output of Indicates l Neurons in a layer network i The gradient of the input, represents the loss function of connection strength, , A represents the intermediate quantity along the spatial direction, B represents the intermediate quantity along the time direction, All represent preset adjustable parameters. Indicates t This iteration l Neurons in a layer network i The membrane potential before output, Indicates t This iteration l Neurons in a layer network i The membrane potential after output, represents the upper threshold of membrane potential, Represents the parameter that controls the temporal variation of the neuron's membrane potential.

[0049] Optionally, the gradient of the connection strength between the above neurons is specifically: ; In the formula, Indicates l Neurons in a layer network i With neurons j The gradient of the connection strength between Indicates l Neurons in a layer network i The gradient of the output of All represent preset adjustable parameters. Indicates t This iteration l Neurons in a layer network i Input, Indicates t This iteration l Neurons in a layer network i The membrane potential after output, Represents the upper threshold of membrane potential.

[0050] S323, based on the gradient of the connection strength between each neuron, the connection strength between each neuron is updated, and after the connection strength between each neuron is updated, the preset adjustable parameters are readjusted; and the above-mentioned training end condition can be to set the maximum number of iterations in advance, and each iteration is executed through the above-mentioned steps S321-S323.

[0051] The above updates the connection strength between each neuron, specifically: ; In the formula, No. l Neurons in a layer network i With neurons j The change in the connection strength between them after the update, represents the learning rate of the model, Indicates l Neurons in a layer network i With neurons j The gradient of the connection strength between A loss function representing the strength of the connection.

[0052] Optionally, the above preset adjustable parameters are specifically: , ; In the formula, All represent preset adjustable parameters. Indicates preset adjustable parameters The minimum value of can be 1. Indicates preset adjustable parameters The maximum value can be 10. T represents the maximum number of iterations, t Indicates the current iteration number.

[0053] Specifically, the neural network in the preset initial model has a multi-layer structure. Each iteration process is to adjust the various model parameters of the model, including the connection strength, etc. after the change value is input into the preset initial model. This can ensure that more neurons participate in learning in the feature learning device, so as to achieve higher accuracy after learning.

[0054] Example 2: The embodiment of the present application provides the application of an online monitoring and suppression system for corrosion of air-conditioning water circuits in valve halls of cold converter stations as described in any one of Example 1 in air-conditioning water circuits.

[0055] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. An online monitoring and suppression system for corrosion of air conditioning waterways in valve halls of cold converter stations, characterized in that: include: An ethylene glycol concentration management module, the ethylene glycol concentration management module includes an ethylene glycol concentration monitoring unit and an ethylene glycol filling unit, wherein: the ethylene glycol concentration monitoring unit is used to obtain the concentration of ethylene glycol in the air conditioning water circuit; the ethylene glycol filling unit is used to add a first target amount of ethylene glycol to the air conditioning water circuit; A corrosion inhibitor management module, the corrosion inhibitor management module includes a pH value monitoring unit and a corrosion inhibitor filling unit, wherein: the pH value monitoring unit is used to obtain the pH value of the solution in the air-conditioning water channel; the corrosion inhibitor filling unit is used to add a second target amount of corrosion inhibitor to the air-conditioning water channel; An ion concentration management module, the ion concentration management module includes an iron ion concentration monitoring unit and an oxygen content monitoring unit, wherein: the iron ion concentration monitoring unit is used to obtain the iron ion content in the air conditioning water path, and the oxygen content monitoring unit is used to obtain the oxygen content of the solution in the air conditioning water path; The operation and maintenance data analysis module is used to determine whether ethylene glycol needs to be added according to the concentration of the ethylene glycol, and calculate the first target amount when the first judgment result is that addition is required, and to determine whether a corrosion inhibitor needs to be added according to the pH value of the solution, and calculate the second target amount when the second judgment result is that addition is required; and is also used to determine the corrosion judgment result of the degree of corrosion of the air-conditioning water channel according to the iron ion content and the oxygen content of the solution.

2. The online monitoring and suppression system for corrosion of air conditioning waterways in valve halls of cold converter stations according to claim 1 is characterized in that: The operation and maintenance data analysis module includes a target quantity calculation unit, and the target quantity calculation unit is used to calculate the first target quantity and the second target quantity, specifically: When the first judgment result is that ethylene glycol needs to be added, a first difference is calculated based on the concentration of the ethylene glycol corresponding to the first judgment result and a preset ethylene glycol standard concentration, and the first difference is input into the ethylene glycol prediction model for processing to obtain the first target amount; When the second judgment result is that corrosion inhibitor needs to be added, the pH value of the solution corresponding to the second judgment result is calculated with the preset standard pH value to obtain a second difference, and the second difference is input into the corrosion inhibitor prediction model for processing to obtain the second target amount.

3. The online monitoring and suppression system for corrosion of air conditioning water channels in valve halls of cold converter stations according to claim 2 is characterized in that: The ethylene glycol prediction model or the corrosion inhibitor prediction model is obtained by the following method: Obtaining the corresponding addition amount, current value, and actual value after addition for each time in the historical addition record, wherein the historical addition record is an ethylene glycol addition record or a corrosion inhibitor addition record; A change value is calculated based on the current value corresponding to each time and the actual value after addition, and the preset initial model is trained using the change value corresponding to each time and the addition amount until the preset initial model reaches the training end condition, and the preset initial model that reaches the training end condition is determined as the ethylene glycol prediction model or the corrosion inhibitor prediction model.

4. The online monitoring and suppression system for corrosion of air conditioning water channels in valve halls of cold converter stations according to claim 3 is characterized in that: During the training process of the preset initial model, each iteration is specifically as follows: After the change value is input into the preset initial model, the output of each neuron in each layer of the network in the preset initial model is calculated; Based on the output of each neuron and the preset adjustable parameters, the gradient of the output of each neuron is calculated, and based on the gradient of the output of each neuron and the preset adjustable parameters, the gradient of the connection strength between each neuron is calculated; Based on the gradient of the connection strength between each neuron, the connection strength between each neuron is updated, and after the connection strength between each neuron is updated, the preset adjustable parameter is readjusted.

5. The online monitoring and suppression system for corrosion of air conditioning water channels in valve halls of cold converter stations according to claim 4 is characterized in that: The output of each neuron in each layer of the network in the preset initial model is as follows: ； In the formula, Indicates t The first iteration l Neurons in a layer network i The output, Indicates l Neurons in a layer network i With neurons j The connection strength between represents the activation function, and the neuron i After output ， ,otherwise ; represents the parameter that controls the change of neuronal membrane potential over time, Indicates t This iteration l Neurons in a layer network i The membrane potential before output, Indicates t This iteration l Neurons in a layer network i Input, represents the upper threshold of membrane potential, Represents the Heaviside function.

6. The online monitoring and suppression system for corrosion of air conditioning waterways in valve halls of cold converter stations according to claim 5 is characterized in that: The gradient of the output of each neuron is specifically: ,in: ， ； In the formula, Indicates l Neurons in a layer network i The gradient of the output of Indicates l- Neurons in a 1-layer network i The gradient of the output of Indicates l Neurons in a layer network i The gradient of the input, represents the loss function of connection strength, , A represents the intermediate quantity along the spatial direction, B represents the intermediate quantity along the time direction, All represent preset adjustable parameters. Indicates t This iteration l Neurons in a layer network i The membrane potential before output, Indicates t This iteration l Neurons in a layer network i The membrane potential after output, represents the upper threshold of membrane potential, Represents the parameter that controls the temporal variation of the neuron's membrane potential.

7. The online monitoring and suppression system for corrosion of air conditioning waterways in valve halls of cold converter stations according to claim 6 is characterized in that: The gradient of the connection strength between neurons is: ； In the formula, Indicates l Neurons in a layer network i With neurons j The gradient of the connection strength between Indicates l Neurons in a layer network i The gradient of the output of All represent preset adjustable parameters. Indicates t This iteration l Neurons in a layer network i Input, Indicates t This iteration l Neurons in a layer network i The membrane potential after output, Represents the upper threshold of membrane potential.

8. The online monitoring and suppression system for corrosion of air conditioning waterways in valve halls of cold converter stations according to claim 7 is characterized in that: Update the connection strength between each neuron, specifically: ； In the formula, No. l Neurons in a layer network i With neurons j The change in the connection strength between them after the update, represents the learning rate of the model, Indicates l Neurons in a layer network i With neurons j The gradient of the connection strength between A loss function representing the strength of the connection.

9. The online monitoring and suppression system for corrosion of air conditioning waterways in valve halls of cold converter stations according to claim 8 is characterized in that: The preset adjustable parameters are specifically: ， ； In the formula, All represent preset adjustable parameters. Indicates preset adjustable parameters The minimum value of Indicates preset adjustable parameters The maximum value of T represents the maximum number of iterations, t Indicates the current iteration number.

Images