A condenser water level vacuum measurement compensation method and measurement device

Abstract

The invention relates to the technical field of thermal power units, and discloses a condenser water level vacuum measurement and compensation method, the method comprising: processing historical data of water level vacuum parameters within a preset time interval based on multivariate state estimation to generate a compensation training model; determining a predicted water level vacuum parameter according to the compensation training model; obtaining a measured water level vacuum parameter, and generating a compensated water level vacuum parameter according to the difference between the predicted water level vacuum parameter and the measured water level vacuum parameter; and also proposes a condenser water level vacuum measurement device, comprising a measuring cylinder connected to the condenser hot well up and down, and also being provided with a vacuum degree measurement device, a water level height measurement device and an exhaust thermometer to measure the water level vacuum parameter, thereby solving the error problem in measurement and the problem of poor anti-infection ability, and improving the measurement accuracy.

Description

Technical Field

[0001] The invention relates to the technical field of thermal power unit measurement, in particular to a condenser water level vacuum measurement compensation method and a measurement device thereof. Background Art

[0002] Condenser vacuum is an important indicator of turbine operation and a major assessment indicator of the comprehensive performance of the condenser. It affects the safe and economic operation of the unit. The vacuum level of the condenser has a direct impact on the economy of the turbine generator unit. For example, if the vacuum of the unit drops by 1%, the heat consumption of the unit will increase by 0.6% to 1%. Therefore, maintaining the vacuum of the condenser is an important guarantee for the economic operation of the unit and an important part of energy conservation in every power plant. Low condenser vacuum has the following hazards: (1) The economy of the unit decreases and the output of the unit is reduced; (2) The exhaust cylinder and the bearing seat expand due to heat, which may cause the center to change and generate vibration; (3) Excessive exhaust temperature may cause the titanium tube of the condenser to loosen and damage the tightness; (4) The axial thrust of the turbine increases; (5) The volume flow of the exhaust steam decreases, which is not conducive to the operation of the last few stages of blades. The last stage will produce flow separation and swirl, and at the same time, it will generate a large exciting force at a certain part of the blade, which may damage the blade and cause an accident.

[0003] Condenser hot well storage - quantitative condensate to ensure that the condensate pump will not be cut off immediately when the load is dumped. Generally, the water level of the condenser hot well should be maintained between 1 / 3 and 2 / 3 of the well. If the condenser water level is too high, part of the condenser tube bundle will be submerged, the space for turbine exhaust condensation will be reduced, the heat exchange space will be reduced, the exhaust temperature will increase, the vacuum will decrease, and the economy of the unit will decrease. If the water level is too low, the condensate pump consumes less power, but it is easy to cause the pump to vaporize, causing serious damage to the impeller, and causing certain vibrations and outlet pressure swings in the pump during operation. No matter which of the above abnormal conditions occurs, it will directly affect the safety and economy of the unit.

[0004] Therefore, for the safe operation of the unit, it is necessary to accurately measure the vacuum degree and water level in the condenser to keep the vacuum degree and water level in the condenser within a safe range. Summary of the invention

[0005] In some embodiments of the present application, in order to solve the above-mentioned technical problems, a condenser water level vacuum measurement compensation method and a measuring device are provided. The measuring device includes a condenser hot well and a measuring tube connected to the condenser hot well. The measuring tube is connected to the gas through an upper connecting pipe and is connected to the water level through a lower connecting pipe. The parameters in the condenser can be obtained by measuring the water level data and vacuum data in the measuring connection. In the measurement compensation method, a compensation training model is generated based on multivariate state estimation. By obtaining compensation data, the error problem in the measurement is solved and the measurement accuracy is improved.

[0006] In some embodiments of the present application, a condenser water level vacuum measurement compensation method is disclosed, the method comprising: processing historical data of water level vacuum parameters within a preset time interval based on multivariate state estimation to generate a compensation training model; determining predicted water level vacuum parameters based on the compensation training model; obtaining measured water level vacuum parameters, and generating compensated water level vacuum parameters based on the difference between the predicted water level vacuum parameters and the measured water level vacuum parameters.

[0007] In some embodiments of the present application, historical data of water level vacuum parameters within a preset time interval are processed based on multivariate state estimation to generate a compensation training model, specifically: the time interval includes a first moment, and the first moment includes multiple variables of water level vacuum parameters, and the variables are recorded as observation vectors, that is:

[0008] X(i)=[x 1 x 2 …x n ] T

[0009] Wherein, i is the first moment, and n is the number of the variables;

[0010] In the first time interval, the observation vectors under different working conditions or at different times are obtained and a process memory matrix is ​​generated, which is:

[0011]

[0012] Among them, m is the number of working conditions or moments, and D is the process memory matrix.

[0013] In some embodiments of the present application, the predicted water level vacuum parameter is determined based on the compensation training model, specifically: the process memory matrix and the measured water level vacuum parameter are compared, and the most similar value of the measured water level vacuum parameter is found in the process memory matrix, and the most similar value is the predicted water level vacuum parameter.

[0014] In some embodiments of the present application, the predicted water level vacuum parameter is determined according to the compensation training model, specifically: the process memory matrix and the measured water level vacuum parameter are obtained and normalized, that is:

[0015]

[0016] Among them, X max , X min are the measured water level vacuum parameters X iThe upper and lower bounds of X are obtained, and X is the value obtained after normalization; the similarity matrix is ​​obtained by combining the normalized value and the process memory matrix; and the predicted water level vacuum parameter is obtained according to the similarity matrix.

[0017] In some embodiments of the present application, after obtaining the predicted water level vacuum parameter, it includes: acquiring the predicted water level vacuum parameter and the measured water level vacuum parameter, and calculating the residual between the predicted water level vacuum parameter and the measured water level vacuum parameter; if the residual is outside the range of the preset accurate residual, regenerating the process memory matrix.

[0018] In some embodiments of the present application, the water level vacuum parameter and the predicted water level vacuum parameter include a water level height value, a vacuum value and an exhaust steam temperature.

[0019] In some embodiments of the present application, a condenser water level vacuum measuring device is also proposed, including: a calculation unit, in which a condenser water level vacuum measurement compensation method is configured; a condenser and a condenser hot well; a measuring cylinder, which is connected to the condenser hot well; a measuring device, which is electrically connected to the calculation unit and transmits the measurement result to the calculation unit; an upper connecting pipe, which connects the upper part of the condenser hot well and the upper part of the measuring cylinder to connect the gas between the condenser hot well and the measuring cylinder; and a lower connecting pipe, which connects the lower part of the condenser hot well and the lower part of the measuring cylinder to connect the water levels of the condenser hot well and the measuring cylinder.

[0020] In some embodiments of the present application, the measuring device includes: a vacuum measuring device, which is connected to the measuring cylinder to measure the vacuum degree of the measuring cylinder; and a water level measuring device, which is connected to the measuring cylinder to measure the water level of the measuring cylinder.

[0021] In some embodiments of the present application, the water level measuring device is configured as a high-frequency radar level gauge.

[0022] The beneficial effects of the present invention are:

[0023] The condenser water level vacuum measuring device and measurement compensation method realize the detection of the water level vacuum of the condenser heat well in an all-in-one manner. At the same time, according to the dynamic measurement value of the water level vacuum, the two parameters can compensate each other, with strong anti-interference ability, thus improving the measurement accuracy and laying the foundation for the safe and economical operation of the unit. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 Schematic diagram of a condenser water level vacuum measuring device in an embodiment of the present invention;

[0025] Figure 2 The figure is a flow chart of a condenser water level vacuum measurement and compensation method in an embodiment of the present invention.

[0026] In the figure: 100, condenser hot well; 200, upper connecting pipe; 300, lower connecting pipe; 400, upper valve; 500, lower valve; 600, measuring tube; 610, vacuum gauge; 620, high-frequency radar level gauge; 630, exhaust steam thermometer; 900, calculation unit. DETAILED DESCRIPTION

[0027] The specific implementation of the present invention is further described in detail below in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0028] In the description of the present application, it should be understood that the terms "center", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present application.

[0029] The terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of this application, unless otherwise specified, "plurality" means two or more.

[0030] In the description of this application, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0031] In some embodiments of the present application, Figure 1 As shown, a condenser water level vacuum measuring device is proposed, which includes a calculation unit 900, a condenser and a condenser hot well, a measuring tube 600, a measuring device, an upper connecting pipe 200 and a lower connecting pipe.

[0032] In some embodiments of the present application, the vacuum degree and water level in the condenser and the condenser hot well are data that need to be detected, and a measuring tube 600 is provided to be connected to the condenser hot well, including an upper connecting pipe 200 connecting the upper part of the condenser hot well and the lower part of the measuring device, so that the gas of the condenser hot well and the measuring tube 600 are connected, and a lower connecting pipe connects the lower part of the condenser hot well and the lower part of the measuring tube 600, so that the water level of the condenser hot well and the measuring tube 600 are connected.

[0033] Therefore, by setting up upper and lower connecting pipes, the vacuum degree and water level in the measuring tube 600 are made the same as the data in the condenser hot well. The data in the condenser hot well can be measured only by the parameters in the measuring tube 600, which facilitates the installation of the measuring device.

[0034] In some specific embodiments, the measuring tube 600 is a cylinder with both ends of a seamless steel pipe with a diameter of φ200 to φ300 mm sealed and welded; one end of the upper connecting tube 200 extends into the upper part of the condenser hot well by 30 to 40 mm, and one end of the lower connecting tube extends into the bottom of the condenser hot well by 40 to 50 mm.

[0035] In some embodiments of the present application, the measuring device also includes a computing unit 900 and a measuring device, the computing unit 900 is configured with a condenser water level vacuum measurement compensation method, the measuring device is electrically connected to the computing unit 900, and the measured data is transmitted to the computing unit 900, and the computing unit 900 performs calculations and controls.

[0036] The measuring device includes a vacuum measuring device, a water level measuring device and an exhaust thermometer 630, wherein the exhaust thermometer 630 is arranged at the exhaust port of the condenser to directly measure the accurate temperature, and the vacuum measuring device and the water level measuring device are arranged at the measuring tube 600. By measuring the vacuum degree and the water level of the measuring tube 600, the vacuum degree and the water level of the condenser hot well can be obtained.

[0037] In some specific embodiments of the present application, the vacuum degree measuring device is a vacuum gauge 610, and the vacuum gauge 610 uses an imported diffused silicon or ceramic core as a pressure detection element. The vacuum gauge 610 is installed at the upper end of the pressure pipe of the vacuum gauge 610, and the lower end of the pressure pipe of the vacuum gauge 610 extends into the measuring tube 600 from the top connecting port of the measuring tube 600.

[0038] In some specific embodiments of the present application, the water level height measuring device is a high-frequency radar level meter 620. The design of the high-frequency radar level meter 620 is based on the principle of time domain reflection, and is mainly composed of a transmitting and receiving device, a signal processor, an antenna, an operation panel, a display, a fault alarm, and the like. The radar level antenna transmits a 26GHz high-frequency microwave pulse, which is transmitted downward through the antenna. After the microwave contacts the surface of the measured medium, it is reflected back and received by the antenna system again. The echo signal is analyzed and processed by the electronic microprocessor and converted into real-time liquid level information represented by an analog quantity of 4 to 20 mA with the help of software.

[0039] The measurements of the vacuum gauge 610 and the high-frequency radar level gauge 620 are in-situ measurements, which are highly real-time and relevant. This feature is used to perform intelligent compensation of the measurement data to improve the measurement accuracy and anti-interference ability of the condenser water level and vacuum.

[0040] In some embodiments of the present application, the measuring device also includes an upper valve 400 and a lower valve 500, which are respectively the first valve of the upper connecting pipe 200, the second valve of the upper connecting pipe 200, the first valve of the lower connecting pipe and the second valve of the lower connecting pipe, and are respectively used to control the connection between the upper connecting pipe 200 and the lower connecting pipe.

[0041] In some embodiments of the present application, a condenser water level vacuum measurement compensation method is also disclosed, which is used to add compensation data to the measurement result when an error occurs in the measurement, so as to obtain a more accurate measurement result.

[0042] The measurement compensation method includes: step S101, processing historical data of water level vacuum parameters within a preset time interval based on multivariate state estimation to generate a compensation training model.

[0043] Specifically, multivariate state estimation is an intelligent pattern recognition technology that realizes state estimation by measuring the similarity between signals under normal conditions. It utilizes normal historical data and learns the relationship between various parameters that can define the system state. By comparing the normal operating state stored in the system with the current observation state, it calculates an estimate of the parameter state of the current measuring device. For new observation vectors of the measuring device, multivariate state estimation predicts the actual parameter state of the measuring device from the learned pattern to participate in intelligent compensation. Therefore, in this embodiment, the historical data of water level vacuum parameters within a time interval of normal operation is learned and the relationship between parameters is defined, and a compensation training model containing the above information is generated to realize the prediction of measurement data.

[0044] In order to generate a compensation training model with high prediction accuracy, in some embodiments of the present application, generating the compensation training model is specifically: within the time interval, obtaining the observation vectors under different working conditions or at different times and generating a process memory matrix.

[0045] Specifically, the time interval includes the first moment, and the water level vacuum parameter at the first moment includes multiple variables, such as vacuum degree, water level height, temperature and other parameters. These variables are recorded as observation vectors, and the public expression is:

[0046] X(i)=[x 1 x 2 … x n ] T

[0047] Among them, i is the first moment and n is the number of variables.

[0048] In the time interval, the observation vectors of different working conditions or multiple moments are obtained to generate the process memory matrix, which is expressed as follows:

[0049]

[0050] Among them, m is the number of working conditions, and D is the process memory matrix.

[0051] Thus, the obtained process memory matrix D is a preliminary compensation training model, which includes water level vacuum parameters under multiple moments and multiple working conditions within a normal working time interval.

[0052] In some embodiments of the present application, in step S102, a predicted water level vacuum parameter is determined according to the compensation training model.

[0053] The specific compensation training model includes various water level vacuum parameters under various working conditions during normal working periods, and the water level vacuum parameters change over time. Therefore, the learning ability of the compensation training model can be used to predict the normal value of the measurement.

[0054] In order to obtain accurate predicted water level vacuum parameters, in some embodiments, the predicted water level vacuum parameters are determined according to the compensation training model, specifically:

[0055] The process memory matrix and the measured water level vacuum parameter are compared, and the most similar value of the measured water level vacuum parameter is found in the process memory matrix, and the most similar value is the predicted water level vacuum parameter.

[0056] It should be noted that the new measured water level vacuum parameters and the historical measured water level vacuum parameters in the process memory matrix are reflected by calculating the Euclidean distance between them. Assuming that the new measured water level vacuum parameters are most similar to the historical measured water level vacuum parameters in the process memory matrix, the Euclidean distance between the two is the smallest, and the most similar historical measured water level vacuum parameters are the predicted water level vacuum parameters.

[0057] In order to obtain accurate predicted water level vacuum parameters, in some embodiments, the predicted water level vacuum parameters are determined according to the compensation training model, and the calculation process is specifically as follows:

[0058] The process memory matrix and the measured water level vacuum parameter are obtained and normalized, that is:

[0059]

[0060] Among them, X max , X min are the measured water level vacuum parameters X i The upper and lower bounds of , X is the value obtained after normalization;

[0061] Combining the normalized values ​​with the process memory matrix to obtain a similarity matrix;

[0062] The predicted water level vacuum parameter is obtained according to the similarity matrix.

[0063] It should be noted that the normalization process is to map all data to 0-1. By normalizing the measured water level vacuum parameters and the process memory matrix, the similarity matrix obtained includes the similarity between each vector and the measured water level vacuum parameters. The vector most similar to the measured water level vacuum parameters is found and then denormalized again. The vector is the predicted water level vacuum parameter.

[0064] In some embodiments of the present application, after obtaining the predicted water level vacuum parameter, it includes: acquiring the predicted water level vacuum parameter and the measured water level vacuum parameter, and calculating the residual between the predicted water level vacuum parameter and the measured water level vacuum parameter; if the residual is outside the range of the preset accurate residual, regenerating the process memory matrix.

[0065] Specifically, the compensation training model improves the prediction accuracy in continuous learning. The accuracy of the compensation training model is measured by the residual between the predicted water level vacuum parameter and the measured water level vacuum parameter. The residual sets a preset accurate residual range. If the residual is within the preset accurate range, it is determined that the vector matrix in the current training model is sufficient to predict the accurate predicted water level vacuum parameter. If the residual is outside the preset accurate residual range, the measured water level vacuum parameter is used as historical data to regenerate the process memory matrix.

[0066] In some embodiments of the present application, step S103 is included to obtain a measured water level vacuum parameter, and generate a compensated water level vacuum parameter according to a difference between the predicted water level vacuum parameter and the measured water level vacuum parameter.

[0067] Specifically, after obtaining accurate predicted water level vacuum parameters, compensated water level vacuum parameters are generated according to the difference between the predicted water level vacuum parameters and the measured water level vacuum parameters to compensate for the error between the measured water level vacuum parameters and the actual data during measurement, thereby improving the measurement accuracy and anti-interference ability.

[0068] In some embodiments of the present application, a condenser water level vacuum measurement compensation method is proposed, the steps of the method are:

[0069] S201, obtaining historical data of water level vacuum parameters within a preset time interval, generating a process memory matrix, wherein the historical data includes observation vectors at multiple moments and under different working conditions, and the observation vector at each moment or each working condition includes multiple variables; executing S202.

[0070] S202, obtaining and comparing the process memory matrix and the measured water level vacuum parameter, and executing S203;

[0071] S203, data processing, normalizing the process memory matrix and the measured water level vacuum parameters, and executing S204;

[0072] S204, generate a similarity matrix, obtain predicted water level vacuum parameters, and execute S205;

[0073] S205, calculating the residual between the predicted water level vacuum parameter and the measured water level vacuum parameter, if the residual is outside the range of the preset accurate residual, executing S201; if the residual is within the range of the preset accurate residual, executing S206;

[0074] S206: Obtain the measured water level vacuum parameter, and generate the compensated water level vacuum parameter according to the difference between the predicted water level vacuum parameter and the measured water level vacuum parameter.

[0075] Those skilled in the art can understand that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention is described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions recorded in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A condenser water level vacuum measurement compensation method, It is characterized in that The method comprises: Processing historical data of water level vacuum parameters within a preset time interval based on multivariate state estimation to generate a compensation training model; Determining predicted water level vacuum parameters according to the compensation training model; Acquire a measured water level vacuum parameter, and generate a compensated water level vacuum parameter according to a difference between the predicted water level vacuum parameter and the measured water level vacuum parameter; Based on multivariate state estimation, the historical data of water level vacuum parameters within a preset time interval are processed to generate a compensation training model, specifically: The time interval includes a first moment, and the first moment includes multiple variables of water level vacuum parameters, and the variables are recorded as observation vectors, namely: Wherein, i is the first moment, and n is the number of the variables; In the time interval, the observation vectors under different working conditions or at different times are obtained and a process memory matrix is ​​generated, which is: Where m is the number of working conditions or moments, and D is the process memory matrix; According to the compensation training model, the predicted water level vacuum parameters are determined, specifically: Compare the process memory matrix and the measured water level vacuum parameter, find the most similar value of the measured water level vacuum parameter in the process memory matrix, and the most similar value is the predicted water level vacuum parameter; After obtaining the predicted water level vacuum parameter, the following steps are performed: Acquiring the predicted water level vacuum parameter and the measured water level vacuum parameter, and calculating the residual between the predicted water level vacuum parameter and the measured water level vacuum parameter; If the residual is outside the range of the preset accurate residual, regenerating the process memory matrix; The water level vacuum parameter, the predicted water level vacuum parameter or the measured water level vacuum parameter includes a water level height value, a vacuum value and an exhaust steam temperature.

2. The compensation method according to claim 1, It is characterized in that According to the compensation training model, the predicted water level vacuum parameters are determined, specifically: The process memory matrix and the measured water level vacuum parameter are obtained and normalized, that is: ； in, , are the measured water level vacuum parameters The upper and lower bounds of is the value obtained after normalization; Combining the normalized values ​​with the process memory matrix to obtain a similarity matrix; The predicted water level vacuum parameter is obtained according to the similarity matrix.

3. A condenser water level vacuum measuring device, It is characterized in that include: A calculation unit, wherein the calculation unit is configured with the condenser water level vacuum measurement compensation method according to any one of claims 1 to 2; Condensers and condenser hot wells; A measuring cylinder, the measuring cylinder being in communication with the condenser hot well; A measuring device, the measuring device is electrically connected to the computing unit and transmits the measurement result to the computing unit; An upper connecting pipe, wherein the upper connecting pipe connects the upper part of the condenser hot well and the upper part of the measuring cylinder, so that the gas of the condenser hot well and the measuring cylinder are connected; A lower connecting pipe is connected with the lower part of the condenser hot well and the lower part of the measuring cylinder to connect the water levels of the condenser hot well and the measuring cylinder.

4. The measuring device according to claim 3, It is characterized in that The measuring device comprises: A vacuum degree measuring device, the vacuum degree measuring device is connected to the measuring cylinder to measure the vacuum degree of the measuring cylinder; A water level measuring device is connected to the measuring cylinder to measure the water level of the measuring cylinder.

5. The measuring device according to claim 4, It is characterized in that The water level measuring device is configured as a high-frequency radar level gauge.

6. The measuring device according to claim 3, It is characterized in that Also includes: An upper valve and a lower valve, wherein the upper valve is arranged on the upper connecting pipe, and the lower valve is arranged on the lower connecting pipe.

Images