Corrosion diagnostic device, corrosion diagnostic method, and corrosion diagnostic program

Abstract

To provide a corrosion diagnostic device, a corrosion diagnostic method, and a corrosion diagnostic program that accurately measure the state of thinning occurring due to local corrosion of a metal facility.SOLUTION: A magnetic sensor array 101 has a plurality of magnetic sensors 111 that are two-dimensionally arranged near a metal member. A filtering processing unit 202 acquires measurement data from the magnetic sensor array 101, and executes filtering processing of filtering the measurement data in a time direction and a spatial direction to generate post-filtering data. A depth measurement processing unit 203 measures corrosion of the metal member on the basis of the post-filtering data calculated by the filtering processing unit 202.SELECTED DRAWING: Figure 2

Description

Technical Field

[0001] The present invention relates to a corrosion diagnosis apparatus, a corrosion diagnosis method, and a corrosion diagnosis program.

Background Art

[0002] In plants such as those in the oil and petrochemical industries, when corrosion, particularly local corrosion, occurs in equipment such as metal pipes, reactors, and distillation towers, it may cause wall thinning in the equipment, resulting in leakage or a decrease in production efficiency. Therefore, there is a need for a technology to detect, measure, and manage the wall thinning caused by locally occurring corrosion and the like in equipment such as metal pipes laid throughout the plant and huge reactors.

[0003] As a technology for detecting, measuring, and managing the wall thinning caused by locally occurring corrosion and the like, a corrosion management system has been proposed. The corrosion management system is composed of, for example, a corrosion inspection apparatus that detects the occurrence of wall thinning caused by corrosion and the like, and a corrosion inspection management apparatus that manages the inspection results.

[0004] Conventionally, as a corrosion inspection apparatus, a technique has been proposed in which an alternating current is passed through a carbon steel pipe and the depth of wall thinning is measured by observing the leakage magnetic flux generated from the pipe surface (for example, Patent Document 1). In this technique, the depth of wall thinning is measured by measuring the leakage magnetic flux from the pipe surface that changes over time with a time sensor array arranged around the pipe. Also, noise filtering based on frequency characteristics is disclosed as a filtering technique for noise removal.

[0005] In addition, as another corrosion detection device, a technique has been proposed in which a carbon steel pipe is magnetized with a permanent magnet and the depth of wall thinning is measured by observing the leakage magnetic flux generated from the pipe surface (for example, Patent Documents 2 to 4). By using this technique, leakage magnetic flux can be generated without passing an electric current through the pipe. According to this technique, when corrosion occurs in the magnetized pipe and the volume decreases, leakage magnetic flux is generated from the pipe surface. By detecting this leakage magnetic flux with a magnetic sensor array, it becomes possible to detect wall thinning and measure the depth.

Prior Art Documents

Patent Documents

[0006]

Patent Document 1

Patent Document 2

Patent Document 3

Patent Document 4

Summary of the Invention

Problems to be Solved by the Invention

[0007] However, in conventional corrosion detection devices, it is considered technically difficult to reduce noise by filtering. For example, the data measured by a sensor array is given as a two-dimensional array. In a corrosion detection device, since the sensors in the sensor array are arranged at intervals, the measured data becomes a two-dimensional signal with coarse resolution. As a method for reducing the noise generated in such data, a method of applying a spatial filter used in image processing can be considered. However, in this method, since there are gaps in the spatial arrangement intervals of the sensors in the sensor array, noise components cause aliasing in the spatial frequency, and the signal components to be extracted and the noise components coexist in the spatial frequency components. Specifically, due to aliasing, high-frequency noise components are not captured and are detected as low-frequency components. Since low-frequency noise is spatially smooth, it is difficult to separate the image of corrosion and the noise. Therefore, it is difficult to apply a noise filter to a two-dimensional signal with coarse resolution. On the other hand, in the case of a technique applying filtering in the time direction, conversely, it is a filtering process that does not consider the continuity of spatial signals. For example, it is assumed that the magnetic flux densities measured by adjacent magnetic sensors are measured to have similar values, but such a relationship is not considered in the filtering process. Therefore, it is difficult to perform appropriate noise filtering on corrosion data having a spatial spread. Therefore, in conventional corrosion detection devices, it has been difficult to accurately measure the state of corrosion.

[0008] An object of the disclosed technology is to provide a corrosion diagnosis device, a corrosion diagnosis method, and a corrosion diagnosis program that accurately measure the state of corrosion caused by local corrosion or the like of metal equipment.

Means for Solving the Problems

[0009] In one aspect of the information processing apparatus and the information processing program disclosed in the present application, the sensor array has a plurality of sensors two-dimensionally arranged in the vicinity of a metal member. The filter processing unit acquires measurement data from the sensor array and performs filter processing of applying a filter in the time direction and the space direction to the measurement data to generate post-filter data. The depth measurement processing unit measures corrosion of the metal member based on the post-filter data calculated by the filter processing unit.

Advantages of the Invention

[0010] On one side, the present invention can accurately measure the state of thinning caused by local corrosion or the like of metal equipment.

Brief Description of the Drawings

[0011]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Modes for Carrying Out the Invention

[0012] Hereinafter, examples of the corrosion diagnosis apparatus, the corrosion diagnosis method, and the corrosion diagnosis program disclosed in the present application will be described in detail with reference to the drawings. Note that the present invention is not limited by this example. Also, the same elements are denoted by the same reference numerals, and redundant descriptions are omitted as appropriate, and each embodiment can be appropriately combined within a non-contradictory range.

[0013] (Embodiment) (Overall Configuration) FIG. 1 is a diagram for explaining an example of the overall configuration of a corrosion diagnosis apparatus according to an embodiment. The corrosion diagnosis apparatus 10 includes a magnetic flux density measurement apparatus 100 and a measurement management apparatus 200. The magnetic flux density measurement apparatus 100 is attached to a metal pipe 11 in the plant 1. The magnetic flux density measurement apparatus 100 and the measurement management apparatus 200 are connected via communication devices 12 and 13 and a network, regardless of whether it is wired or wireless.

[0014] The plant 1 is an example of various plants using oil, petrochemicals, chemicals, gas, etc., and includes factories and the like equipped with various facilities for obtaining products. Examples of products are LNG (liquefied natural gas), resins (plastics, nylon, etc.), chemical products, etc. Examples of facilities are factory facilities, machine facilities, production facilities, power generation facilities, storage facilities, facilities at the wellhead for mining oil, natural gas, etc.

[0015] The inside of the plant 1 is constructed using a distributed control system (DCS) or the like. For example, although not shown, the control system inside the plant 1 uses the process data used in the plant 1 to perform various controls on control devices such as field devices installed in the facilities to be controlled and operation devices corresponding to the facilities to be controlled. The control system includes a computer such as a server.

[0016] (Corrosion Diagnosis Apparatus) The corrosion diagnosis apparatus 10 is an apparatus for diagnosing the corrosion state of the metal pipe 11 arranged in the plant 1. However, the corrosion diagnosis apparatus 10 is not limited to the metal pipe 11, and it is also possible to diagnose the corrosion state of other parts as long as they are metal parts.

[0017] The magnetic flux density measuring device 100 is a device that is attached to the metal pipe 11 and measures the spatial distribution of the magnetic flux density in the space including the inside of the metal pipe 11. The measurement management device 200 is a device that measures the wall thickness reduction depth of the metal pipe 11 using the spatial distribution of the magnetic flux density measured by the magnetic flux density measuring device 100.

[0018] FIG. 2 is a block diagram of the magnetic flux density measuring device and the measurement management device according to the embodiment. Next, with reference to FIG. 2, details of the magnetic flux density measuring device 100 and the measurement management device 200 will be described.

[0019] The magnetic flux density measuring device 100 is equipped with magnetic sensors 111 arranged in an array and measures the leakage magnetic flux generated from the pipe. As shown in FIG. 2, the magnetic flux density measuring device 100 includes a magnetic sensor array 101 and a magnetic flux density measuring unit 102. Further, the magnetic flux density measuring unit 102 includes a magnetic flux density acquisition unit 121 and a magnetic flux density storage unit 122.

[0020] The magnetic sensor array 101 has a magnetic sensor group including a plurality of magnetic sensors 111. The magnetic sensor array 101 is formed by arranging the magnetic sensors 111 in an array. The magnetic sensor array 101 is attached to the metal pipe 11.

[0021] The magnetic sensors 111 are arranged side by side at regular intervals in the extending direction of the metal pipe 11, that is, in the direction along the axis of the metal pipe 11. For example, about 10 magnetic sensors 111 are arranged in the direction along the axis of the metal pipe 11. Also, the magnetic sensors 111 are arranged side by side at regular intervals in the direction perpendicular to the axis of the metal pipe 11. The magnetic sensors 111 may be arranged so as to surround the outer periphery of the metal pipe 11, or may be arranged on a part of the outer periphery of the metal pipe 11.

[0022] The magnetic sensor 111 is composed of a magnetic susceptibility element such as a Hall element, a magnetoresistive effect element, or a magnetic impedance effect element. For example, the magnetic sensor 111 has a cubic shape with each side being about 5 cm. Each magnetic sensor 111 can measure the magnetic fields of three orthogonal axes (for example, the X-axis, Y-axis, and Z-axis). The magnetic sensor 111 can also operate as a sensor for measuring the magnetic field of a single axis or two orthogonal axes.

[0023] The magnetic sensor array 101 measures the spatial distribution of the magnetic flux density based on the measurement results of the magnetic field by each magnetic sensor 111. Then, the magnetic sensor array 101 outputs the measured value of the magnetic flux density as two-dimensional data to the magnetic flux density acquisition unit 121 according to the arrangement of the magnetic sensors 111. For each measurement, the magnetic sensor array 101 outputs three two-dimensional data representing each component of the three orthogonal axes.

[0024] The magnetic flux density acquisition unit 121 receives a command for measuring the magnetic flux density from the measurement command unit 201 of the measurement management device 200, which will be described later, via the communication device 13, the network, and the communication device 12. Then, the magnetic flux density acquisition unit 121 acquires measurement data including the measured value of the magnetic flux density, which is two-dimensional data of the measured values for each component in the three orthogonal axes, from the magnetic sensor array 101. And the magnetic flux density acquisition unit 121 stores the acquired measurement data of the magnetic flux density in the magnetic flux density storage unit 122.

[0025] The magnetic flux density storage unit 122 is a storage device. The magnetic flux density storage unit 122 stores the measurement data of the magnetic flux density acquired by the magnetic flux density acquisition unit 121.

[0026] The communication devices 12 and 13 are devices for transmitting and receiving data. The communication device 12 and the communication device 13 transmit and receive data between the magnetic flux density measurement device 100 and the measurement management device 200 via the network. There is no particular limitation on the communication method used by the communication devices 12 and 13. Also, the communication device 12 and the communication device 13 may be the same type of device or different types of devices.

[0027] Next, the measurement management device 200 will be described. The measurement management device 200 executes processes such as noise removal for the measurement data of the magnetic flux density measured by the magnetic flux density measurement device 100 and calculation of the depth of material removal. The measurement management device 200 may be a device dedicated to magnetic flux density or a server. As shown in FIG. 2, the measurement management device 200 includes a measurement command unit 201, a filter processing unit 202, a depth measurement processing unit 203, and a post-processing unit 204.

[0028] The measurement command unit 201 generates a command for causing the magnetic flux density measurement device 100 to measure the magnetic flux density. The measurement command unit 201 transmits, for example, a command for measuring the magnetic flux density that serves as a trigger for measurement at a specific measurement interval such as once a day to the magnetic flux density acquisition unit 121 of the magnetic flux density measurement device 100.

[0029] The filter processing unit 202 removes noise from the measurement data of the magnetic flux density measured by the magnetic flux density measurement device 100. The filter processing unit 202 includes a filter design value storage unit 221, a measurement data storage unit 222, a filter calculation unit 223, a filter result storage unit 224, and a data generation unit 225.

[0030] The filter design value storage unit 221 is a storage device that stores design values used for filter processing (filtering). The filter design value storage unit 221 stores the design values preset by the user at the configuration stage. For example, the probability density function calculated as a result of performing filter processing on the measurement data by the filter calculation unit 223 described later changes depending on the distribution followed by the noise. Therefore, the filter design value storage unit 221 stores the type of distribution followed by the noise preset by the user as a set value. As the distribution followed by the noise, for example, a normal distribution can be used. In addition, the type of distribution followed by the noise may be any distribution that takes continuous values, and gamma distribution, T distribution, etc. can be used.

[0031] Further, when the type of the preset distribution is a normal distribution, for example, the filter design value storage unit 221 stores a unit matrix used for filter processing, a precision matrix of the prior distribution of the true signal, and the like. In addition, the filter design value storage unit 221 may store a preset variance.

[0032] The measurement data storage unit 222 acquires the measurement data of the magnetic flux density obtained by the magnetic sensor array 101 of the magnetic flux density measuring device 100 from the magnetic flux density storage unit 122. Then, the measurement data storage unit 222 stores and accumulates the acquired measurement data together with the past measurement data. The measurement data storage unit 222 is, for example, a ring buffer and can accumulate a predetermined number of measurement data. When the stored measurement data reaches a predetermined number, the measurement data storage unit 222 deletes the old measurement data in order and stores the newly acquired measurement data. In the present embodiment, the measurement data storage unit 222 acquires the measurement data for a plurality of times from the magnetic flux density storage unit 122 in a lump.

[0033] The filter calculation unit 223 acquires all of the predetermined number of measurement data stored in the measurement data storage unit 222. Then, the filter calculation unit 223 performs filter processing on the acquired measurement data. The filter calculation unit 223 performs filter processing on each component of the three orthogonal axes. At this time, the filter calculation unit 223 may perform processing individually on each component of the three orthogonal axes, or may perform filter processing collectively. Specifically, the filter calculation unit 223 extracts signal components commonly included in a predetermined number of measurement data by a statistical method. The filter calculation unit 223 can avoid noise aliasing in the spatial frequency by using a predetermined number of measurement data that are measurement values for a plurality of times.

[0034] Here, since the predetermined number of measurement data used for the filtering process contains noise, there are variations in the results of filtering the measured values. Therefore, the filter operation unit 223 calculates a probability density function as a result of performing a filtering process on the measurement data using a statistical method. By sampling the measurement data after the filtering process according to the probability density function calculated by the filter operation unit 223, a large number of measurement data after the filtering process can be generated. Details of the filtering process by the filter operation unit 223 will be described below.

[0035] The filter operation unit 223 uses statistical processing for the filtering process. In this embodiment, a method using Bayes' theorem will be described as an example of the filtering process using statistical processing. The filter operation unit 223. For example, the filtering process is performed for each of the measurement data of the components of the three orthogonal axes, or after formulating all the components of the three axes, the filtering process is performed collectively on all the components of the two axes. Hereinafter, the filtering process for one of the three components of the three orthogonal axes will be described.

[0036] The filter operation unit 223 describes, using Bayes' theorem, the probability that the true signal that commonly appears in the measurement data (g1, g2, g3, ···, g4) to be filtered is z as follows. Hereinafter, the measurement data to be filtered is referred to as the data to be filtered.

[0037]

Equation

[0038] Z is a random variable and z is a realized value. The left side of Equation (1) represents the probability that the true signal is z when the measurement data (g1, g2, g3, ···, g M ) is obtained. The first term in the numerator on the right side is the probability that the measured measurement data is (g1, g2, g3, ···, g M) represents the probability. Also, the second term in the numerator on the right side represents the probability that the measurement data is z. The second term in the numerator on the right side is given, for example, from empirical rules. Also, the denominator on the right side represents the probability that the actually measured measurement data is (g1, g2, g3, ···, g M ) is.

[0039] The filter operation unit 223 uses, as the result after the filter process, z such that the probability on the left side in the mathematical formula (1) becomes maximum. Also, since the left side of the mathematical formula (1) is a probability density function, the filter operation unit 223 can express how much the signal after the filter varies according to the mathematical formula (1).

[0040] However, the filter operation unit 223 may estimate the parameters of the probability density function included in the right side of the mathematical formula (1) from the measurement data. For example, it estimates the parameters included in the mathematical formula (1) so that the probability of occurrence of g1~g M becomes maximum. In this case, it is not necessary to previously store the parameters included in the mathematical formula (1) in the filter design value storage unit 221.

[0041] Then, the filter operation unit 223 stores, in the filter result storage unit 224 as the result of the filter process for the measurement data, the parameters of the probability density function followed by the measurement data after the filter process forming the mathematical formula (1).

[0042] The filter result storage unit 224 stores the processing result obtained by the filter process performed by the filter operation unit 223. Since the result of the filter process is output as a probability density function, the filter result storage unit 224 stores the parameters of the probability density function followed by the measurement data after the filter process represented by the mathematical formula (1).

[0043] Also, above, the case where the type of the distribution followed by the noise is a normal distribution has been described as an example. On the other hand, when the type of the distribution followed by the noise is another distribution, for the filter target data (g1, g2, g3, ···, g M) is a distribution represented by a transformation formula when it is assumed to be independent of the time series. The distribution is determined according to the type of distribution. Therefore, when the type of distribution followed by the noise is a different distribution from other distributions, the filter operation unit 223 may determine the probability density function according to the transformation formula distribution when it is assumed that (g1, g2, g3, ···, g M ) is independent of the time series.

[0044] The data generation unit 225 generates the measurement data after the filter process is performed. The data generation unit 225 acquires the parameters of the probability density function followed by the measurement data after the filter process represented by the mathematical formula (1) from the filter result storage unit 224. Then, the data generation unit 225 generates a plurality of measurement data after the filter process using the acquired information. For example, the data generation unit 225 can generate the measurement data after the filter process by sampling data according to the probability density function represented by the mathematical formula (1). Hereinafter, the measurement data after the filter process is referred to as "post-filter data". Then, the data generation unit 225 outputs the post-filter data to the depth measurement processing unit 203.

[0045] As described above, the filter operation unit 223 and the data generation unit 225 execute a filter process for removing noise from the measurement data by calculating the probability density function by statistical processing and generating the post-filter data using the calculated probability density function. Furthermore, the type of distribution followed by the noise can be varied, and the filter operation unit 223 can select the type of statistical processing according to the type of distribution followed by the noise.

[0046] The depth measurement processing unit 203 measures the depth of the wall thickness reduction generated in the metal pipe 11. The depth measurement processing unit 203 includes a depth measurement design value storage unit 231, a depth measurement unit 232, and a depth measurement result storage unit 233.

[0047] The depth measurement design value storage unit 231 stores various design values such as parameters used for calculating the depth measurement of the metal loss. The depth measurement design value storage unit 231 stores values preset by the user at the configuration stage. For example, the depth measurement design value storage unit 231 stores information on the relative positional relationship of each of the magnetic sensors 111, as well as information on the shape of the metal pipe 11 to be subject to corrosion diagnosis and the positional relationship with each magnetic sensor 111.

[0048] The depth measurement unit 232 acquires the filtered data generated by the data generation unit 225 of the filter processing unit 202. The depth measurement unit 232 also acquires various design values used for calculating the depth measurement of the metal loss from the depth measurement design value storage unit 231.

[0049] Then, the depth measurement unit 232 measures the depth of the metal loss in the metal pipe 11 for each filtered data by calculating the measurement value of the depth of the metal loss in the metal pipe 11 for each filtered data using the filtered data and the design value information. The depth measurement unit 232 calculates a plurality of measurement values of the depth for each filtered data. Thereafter, the depth measurement unit 232 stores the measurement value of the depth of the metal loss in the metal pipe 11 in the depth measurement result storage unit 233.

[0050] The depth measurement unit 232 converts the filtered data into the shape and size of the metal loss by the following method. For example, the depth measurement unit 232 calculates the difference between the filtered data calculated based on the magnetic flux density measured by the magnetic sensor array 101 in the state without metal loss and the filtered data calculated from the measurement data, and can convert the shape and size of the metal loss based on the difference. Also, the depth measurement unit 232 can apply the filtered data calculated from the measurement data to a physical model, calculate the magnetic field distribution, current distribution, and resistance distribution, and calculate the metal wall thickness from the resistance distribution.

[0051] The depth measurement result storage unit 233 acquires and stores the measurement value of the depth of the metal loss in the metal pipe 11 calculated by the depth measurement unit 232.

[0052] The post-processing unit 204 performs post-processing such as determining the risk level of wall thinning in the metal pipe 11 and providing information. The post-processing unit 204 includes a post-processing setting storage unit 241, a risk depth determination unit 242, and a display processing unit 243.

[0053] The post-processing setting storage unit 241 stores a determination threshold value for determining whether the depth of wall thinning is a dangerous depth. The post-processing setting storage unit 241 stores the determination threshold value preset by the user at the configuration stage.

[0054] The risk depth determination unit 242 acquires the measured value of the wall thinning depth in the metal pipe 11 from the depth measurement result storage unit 233. Also, the risk depth determination unit 242 acquires a determination threshold value for determining whether the wall thinning depth is a dangerous depth from the post-processing setting storage unit 241. Next, the risk depth determination unit 242 compares the acquired measured value of the wall thinning depth in the metal pipe 11 with the determination threshold value. If the measured value of the wall thinning depth in the metal pipe 11 is equal to or greater than the determination threshold value, it is determined that the wall thinning depth in the metal pipe 11 has reached the dangerous level. Then, the risk depth determination unit 242 sends a notification to the display processing unit 243 that the wall thinning depth in the metal pipe 11 has reached the dangerous level. Also, the risk depth determination unit 242 may send the determination threshold value to the display processing unit 243.

[0055] The display processing unit 243 acquires the measured value of the wall thinning depth in the metal pipe 11 from the depth measurement result storage unit 233. Then, the display processing unit 243 generates a wall thinning depth graph, which is a trend graph with error bars representing the transition of the wall thinning depth in the metal pipe 11.

[0056] FIG. 3 is a diagram showing an example of a meat reduction depth graph. The trend graph with error bars may be, for example, a box-and-whisker plot as shown in FIG. 3. The meat reduction depth graph 300, which is a box-and-whisker plot, represents the maximum depth of meat reduction in the metal pipe 11 on the vertical axis and the measurement date on the horizontal axis. In one depth measurement process, since multiple measurements are performed to obtain the probability density function, the display processing unit 243 uses the variation in the measurement results to express the width 301 in the depth direction on each measurement date and generates error bars. Here, the display processing unit 243 represents the average value of the depth measurement values by points 302 and displays the trend of the average value together with the box-and-whisker plot. In this way, the display processing unit 243 can visualize the reliability of the depth measurement values by using the meat reduction depth graph 300, which is a trend graph with error bars, for the measurement results of the meat reduction depth in the metal pipe 11.

[0057] Also, the display processing unit 243 can obtain the determination threshold value from the dangerous depth determination unit 242. In that case, the display processing unit 243 may display the depth corresponding to the determination threshold value as the dangerous depth on the meat reduction depth graph. For example, as shown in FIG. 3, the display processing unit 243 may add a danger depth display line 303 to the meat reduction depth graph 300, which is a box-and-whisker plot.

[0058] Furthermore, when the display processing unit 243 receives a notification from the dangerous depth determination unit 242 that the depth of meat reduction in the metal pipe 11 has reached the dangerous level, the display processing unit 243 adds an alert notifying that the depth of meat reduction has reached the dangerous level to the meat reduction depth graph.

[0059] The display processing unit 243 visualizes by displaying the generated meat reduction depth graph on a display device such as a monitor. For example, if the corrosion diagnosis device 10 is a dedicated device for performing corrosion diagnosis, the display processing unit 243 may display it on the monitor attached to the corrosion diagnosis device 10. Also, if the corrosion diagnosis device 10 is a server, the display processing unit 243 may embed the generated graph in HTML, send it to the client computer, and display it on the user's browser.

[0060] (Flow of Corrosion Diagnosis Process) Figure 4 is a flowchart of the magnetic flux density measurement process by the magnetic flux density measurement device according to the embodiment. Next, with reference to FIG. 4, the flow of the magnetic flux density measurement process by the magnetic flux density measurement device 100 according to the embodiment will be described.

[0061] The magnetic flux density acquisition unit 121 determines whether it has received a command to measure the magnetic flux density from the measurement command unit 201 of the measurement management device 200 (step S101). If it has not received the command to measure the magnetic flux density (step S101: NO), the magnetic flux density acquisition unit 121 waits until it receives the command to measure the magnetic flux density.

[0062] On the other hand, if it has received the command to measure the magnetic flux density (step S101: YES), the magnetic flux density acquisition unit 121 acquires measurement data of the magnetic flux density, which is two-dimensional data of the measured values for each component in the three orthogonal axes, from the magnetic sensor array 101. Then, the magnetic flux density acquisition unit 121 stores the acquired measurement data of the magnetic flux density in the magnetic flux density storage unit 122. (Step S102).

[0063] Thereafter, the magnetic flux density measurement unit 102 transmits the measurement data stored in the magnetic flux density storage unit 122 to the filter processing unit 202 of the measurement management device 200 via the communication device 12, the network, and the communication device 13 (step S103).

[0064] Figure 5 is a flowchart of the corrosion diagnosis process by the measurement management device according to the embodiment. Next, with reference to FIG. 5, the flow of the corrosion diagnosis process by the measurement management device 200 according to the embodiment will be described.

[0065] The measurement command unit 201 transmits a command to measure the magnetic flux density to the magnetic flux density acquisition unit 121 of the magnetic flux density measurement device 100 at a specific measurement interval (step S201).

[0066] The measurement data storage unit 222 of the filter processing unit 202 receives, stores, and holds the measurement data transmitted from the magnetic flux density measurement unit 102 of the magnetic flux density measurement device 100, and updates it with the newly acquired measurement data (step S202). Here, the case where the measurement data storage unit 222 acquires and holds M pieces of measurement data will be described.

[0067] The filter calculation unit 223 acquires M pieces of measurement data of the magnetic flux density from the measurement data storage unit 222 (step S203). Also, the filter calculation unit 223 acquires design values used for filter processing, such as the type of distribution followed by the noise, from the filter design value storage unit 221.

[0068] The filter calculation unit 223 executes filter processing on the measurement data using Equation (1), and outputs and stores the probability density function followed by the post-filter data in the filter result storage unit 224 (step S204).

[0069] The data generation unit 225 acquires the probability density function followed by the post-filter data from the filter result storage unit 224. Then, the data generation unit 225 generates N pieces of post-filter data using the acquired probability density function (step S205). Thereafter, the data generation unit 225 outputs the N generated post-filter data to the depth measurement unit 232 of the depth measurement processing unit 203.

[0070] Next, the depth measurement unit 232 receives the input of N pieces of post-filter data from the data generation unit 225. Also, the depth measurement unit 232 acquires parameters used for the calculation of the depth measurement of the material removal depth from the depth measurement design value storage unit 231. Then, the depth measurement unit 232 initializes n representing the number for identifying the post-filter data used for the depth measurement and sets n = 1 (step S206).

[0071] Next, the depth measurement unit 232 performs depth measurement using the n-th post-filter data (step S207).

[0072] Next, the depth measurement unit 232 stores the measured depth value obtained by the measurement in the depth measurement result storage unit 233 (step S208).

[0073] Next, the depth measurement unit 232 determines whether n is N or more (step S209). If n is less than N (step S209: NO), the depth measurement unit 232 increments n by 1 (step S210) and returns to step S207.

[0074] On the other hand, if n is N or more (step S209: YES), the danger depth determination unit 242 of the post-processing unit 204 acquires the N depth measurement values that are the measurement results from the depth measurement result storage unit 233. Further, the danger depth determination unit 242 acquires a determination threshold value for determining whether the depth of material removal is a dangerous depth from the post-processing setting storage unit 241. Then, the danger depth determination unit 242 determines whether any of the depth measurement values that are the measurement results has reached the dangerous depth (step S211). If none of the depth measurement values has reached the dangerous depth (step S211: NO), the corrosion diagnosis process proceeds to step S213.

[0075] On the other hand, if any of the depth measurement values has reached the dangerous depth (step S211: YES), the danger depth determination unit 242 notifies the display processing unit 243 that the depth of material removal has reached the dangerous level (step S212).

[0076] The display processing unit 243 acquires the measured value of the depth of material removal in the metal pipe 11 from the depth measurement result storage unit 233. Then, the display processing unit 243 generates a material removal depth graph, which is a trend graph with error bars representing the transition of the depth of material removal in the metal pipe 11. Further, when the display processing unit 243 receives a notification from the danger depth determination unit 242 that the depth of material removal has reached the dangerous level, the display processing unit 243 adds an alert notifying that the depth of material removal has reached the dangerous level to the material removal depth graph. Then, the display processing unit 243 displays the generated material removal depth graph on a display device such as a monitor (step S213).

[0077] Here, in the above description, the metal pipe 11 has been described as an example. However, the corrosion diagnosis apparatus 10 according to the embodiment may not have a cylindrical shape as the diagnosis target. For example, it can also diagnose corrosion occurring on a plate-shaped metal member or a tank wall surface.

[0078] (Effect) As described above, the corrosion diagnosis apparatus according to the embodiment acquires measurement data of magnetic flux density at a plurality of timings, performs filter processing on the acquired plurality of measurement data, and generates a probability density function followed by the post-filter data. Then, the corrosion diagnosis apparatus generates a plurality of post-filter data using the generated probability density function, and generates a thinning depth graph, which is a trend graph with error bars representing the depth of corrosion of the metal part, using the generated plurality of post-filter data, and provides it to the user.

[0079] Thereby, unlike the conventional spatial filter, the corrosion diagnosis apparatus according to the embodiment can add a smoothing process in the time direction to the filter process, and can perform a filter process that reduces the influence of aliasing in the spatial frequency. Further, the corrosion diagnosis apparatus can select whether to use a filter for time series processing or a filter that is a combination of a spatial filter and time series processing depending on whether C, which is the precision matrix of the prior distribution of the true signal included in the filter process, is a diagonal matrix. Furthermore, as a result of the filter process, the corrosion diagnosis apparatus according to the embodiment can calculate a probability density function followed by the post-filter data, and thus can also reflect the variation of noise in subsequent processes. For example, the corrosion diagnosis apparatus according to the embodiment samples a large number of post-filter data according to the probability density function, and performs depth measurement on all the sampled post-filter data. Thereby, the corrosion diagnosis apparatus according to the embodiment can visualize the variation indicating how much error occurs in the depth due to the influence of noise. Based on the visualized information provided by the corrosion diagnosis apparatus, the user can grasp how reliable the result calculated by the corrosion diagnosis apparatus is.

[0080] (System) Regarding the processing procedures, control procedures, specific names, and information including various data and parameters shown in the above-mentioned documents and drawings, they can be arbitrarily changed unless otherwise specified.

[0081] In addition, each component of each illustrated device is a functional concept and does not necessarily have to be physically configured as shown in the figure. That is, the specific forms of distribution and integration of each device are not limited to those shown in the figure. In other words, all or part of it can be functionally or physically distributed and integrated in any unit according to various loads, usage situations, etc.

[0082] For example, all or part of the functions of the measurement management device 200 may be incorporated into the magnetic flux density measurement device 100 equipped with the magnetic sensor array 101. Also, the measurement management device 200 may be a server on the cloud.

[0083] Furthermore, each processing function performed by each device can be realized in whole or in any part by a CPU (Central Processing Unit) and a program analyzed and executed by the CPU, or can be realized as hardware by wired logic.

[0084] (Hardware) Next, a hardware configuration example of the measurement management device 200 will be described. FIG. 6 is a hardware configuration diagram of the measurement management device. As shown in FIG. 6, the measurement management device 200 includes a processor 91, a memory 92, a communication device 93, and an HDD (Hard Disk Drive) 94. Also, the processor 91 is connected to the memory 92, the communication device 93, and the HDD 94 via a bus.

[0085] The communication device 93 is a network interface card or the like and is used for communication with other information processing devices. For example, the communication device 93 relays communication between the processor 91 and maintenance devices such as a control system arranged in the plant 1.

[0086] The HDD 94 is an auxiliary storage device. The HDD 94 realizes the functions of the filter design value storage unit 221, the measurement data storage unit 222, the filter result storage unit 224, the depth measurement design value storage unit 231, the depth measurement result storage unit 233, and the post-processing setting storage unit 241. Also, the HDD 94 stores various programs including a program for realizing the functions of the measurement command unit 201, the filter calculation unit 223, the data generation unit 225, the depth measurement unit 232, the dangerous depth determination unit 242, and the display processing unit 243 illustrated in FIG. 2.

[0087] The processor 91 reads out various programs stored in the HDD 94, expands them in the memory 92, and executes them. Thereby, the processor 91 realizes the functions of the measurement command unit 201, the filter calculation unit 223, the data generation unit 225, the depth measurement unit 232, the dangerous depth determination unit 242, and the display processing unit 243 illustrated in FIG. 2.

[0088] In this way, the measurement management device 200 operates as an information processing device that executes various processing methods by reading out and executing a program. Also, the measurement management device 200 can read out the program from the recording medium by the medium reading device and realize the same functions as those in the above-described embodiments by executing the read program. Note that the program here is not limited to being executed by the measurement management device 200. For example, the present invention can be similarly applied when another computer or server executes the program, or when these cooperate to execute the program.

[0089] This program can be distributed via a network such as the Internet. Also, this program is recorded on a computer-readable recording medium such as a hard disk, a flexible disk (FD), a CD-ROM, a MO (Magneto-Optical disk), a DVD (Digital Versatile Disc), and can be executed by being read out from the recording medium by a computer.

[0090] Furthermore, in the above-described embodiments, the measurement of the depth of pipe corrosion has been described as an example. However, the above-described filter processing can be applied as long as the following conditions are satisfied. The first condition is that the temporal change of the signal is sufficiently slow, or the sampling frequency is sufficiently high with respect to the frequency of the target signal. The second condition is to use the measurement data obtained from the measurement using the sensor array. For example, when measuring the height by arranging a sensor array around a growing plant, it is also possible to measure the height of the plant by removing noise using the filter processing in the embodiment.

Description of Signs

[0091] 1 Plant 10 Corrosion diagnosis device 11 Metal pipe 12, 13 Communication device 100 Magnetic flux density measurement device 101 Magnetic sensor array 102 Magnetic flux density measurement unit 111 Magnetic sensor 121 Magnetic flux density acquisition unit 122 Magnetic flux density storage unit 200 Measurement management device 201 Measurement command unit 202 Filter processing unit 203 Depth measurement processing unit 204 Post-processing unit 221 Filter design value storage unit 222 Measurement data storage unit 223 Filter calculation unit 224 Filter result storage unit 225 Data generation unit 231 Depth measurement design value storage unit 232 Depth measurement unit 233 Depth measurement result storage unit 241 Post-processing setting storage unit 242 Dangerous depth determination unit 243 Display processing unit

Claims

1. A sensor array having a plurality of sensors two-dimensionally arranged in the vicinity of a metal member, a filter processing unit that acquires measurement data by the sensor array and performs a filtering process of applying a filter in the time direction and the space direction to the measurement data to generate post-filter data, a depth measurement processing unit that measures corrosion of the metal member based on the post-filter data calculated by the filter processing unit A corrosion diagnosis apparatus comprising:

2. The corrosion diagnosis apparatus according to claim 1, wherein the filter processing unit applies a filter in the space direction based on individual data acquired by the sensors at different positions, and applies a filter in the time direction based on a plurality of measurement data acquired by the sensor array at different timings.

3. The corrosion diagnosis apparatus according to claim 1 or 2, wherein the filter processing unit performs a filter process based on statistical processing and calculates a probability density function followed by the post-filter data.

4. The corrosion diagnosis apparatus according to claim 3, wherein the type of the statistical processing is selectable.

5. The filter processing unit generates a plurality of the post-filter data based on the probability density function, the depth measurement processing unit measures corrosion of the metal member for each of the post-filter data to obtain a plurality of measurement results, and further comprises a display processing unit that generates and provides a graph including error bars of the measurement results corresponding to noise to be removed by the filter processing based on the plurality of measurement results. The corrosion diagnosis apparatus according to claim 3, characterized in that:

6. The corrosion diagnosis apparatus according to claim 3, further comprising a danger depth determination unit that compares a measurement result by the measurement processing unit with a determination threshold value to determine whether or not the corrosion has reached a danger level and notifies the result.

7. Obtaining measurement data using a sensor array having a plurality of sensors two-dimensionally arranged in the vicinity of a metal member, performing a filtering process of applying a filter in the time direction and the space direction to the measurement data to calculate post-filter data, measuring corrosion of the metal member based on the calculated post-filter data A corrosion diagnosis method characterized by causing a computer to execute the process.

8. Obtaining measurement data using a sensor array having a plurality of sensors two-dimensionally arranged in the vicinity of a metal member, Execute a filtering process of applying a filter in the time direction and the space direction to the measurement data to calculate post-filter data. Based on the calculated post-filter data, measure the corrosion of the metal member. A corrosion diagnosis program characterized by causing a computer to execute the process.

Images