JPS6319230B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION Field of the Invention The present invention is adapted to move a plurality of particles sequentially past at least one detector responsive to a desired property of the particles. The present invention relates to a particle sorting device.

(Prior Art) In a radiometric sorting device, ore particles are caused to flow in parallel, and the particles in each flow are spaced apart from each other.

The particles in each stream are caused to pass over a plurality of spaced apart scintillation detectors, each of which records radiation counts from each passing particle.
Counts from individual detectors for the same particle are accumulated to finalize the radioactive content of the particle, and a decision to accept or reject the particle is made based on this determination.

This method gives reasonable results when the spacing between adjacent particles is large, but as the spacing decreases, the cumulative count obtained from a particle (P) is at least as large as the previous particle (P-1). and the peripheral effects arising from the later particle (P+1).

Due to the continuous and irregular nature of radiation emission from radioactive materials, when particle (P) is within the gated counting region of a particular scintillation detector, particles (P-1) and (P+1) are also radioactive. will be detected and counted by the detector and associated counting electronics as radiation emitted by the particle (P). As a result, if particle (P-1) or (P+
If 1) is a fairly high-grade ore and the particles (P) are waste rock or low-grade ore,
Particles (P) will have an apparently high count and will therefore be improperly sorted by the equipment as high-grade ore, so if it is actually waste rock, it will be considered a high-grade ore of the incoming ore. This results in a decrease in the proportion of This effect cannot be avoided given the particle-to-detector distance required to obtain adequate sensitivity and the particle spacing required to provide commercially acceptable particle delivery rates. . This effect is further supplemented by the effects of particles (P-2) and (P+2), but these are secondary effects and can be ignored.

For example, as a practical matter, the optimum for a particle of 37 mm,
Grade at 100mm intervals before 37mm waste rock particles (P)
If there is a particle (P-1) of 0.5Kg/ton, the particle (P) will be detected as a 0.12Kg/ton particle, so the receiving device set to 0.1Kg/ton will incorrectly accept it. It turns out. This is ignoring the additional influence of the neighboring ore particles, which further increase the apparent quality of the particle (P). This effect increases rapidly as the particles become larger and the particle spacing becomes smaller.

SUMMARY OF THE INVENTION A method of particle sorting provided by the present invention includes sequentially moving a plurality of particles past at least one detector 12 that is responsive to radiation emitted from the particles. , generating for each particle an output signal that depends on the intensity of radiation emitted from the particle by the reaction of the detector, wherein the sorting method determining a spacing between at least one neighboring particle and a calibration factor depending on the spacing and an output signal depending on the intensity of radiation emitted by the particle of interest; The present invention is characterized in that it uses a particle measurement method that includes the step of providing an output signal.

Further, in the present invention, the particles are moved sequentially past a plurality of detectors, and the output signal for each particle provides a separate radiation count of the detectors for the particle. It is designed to be generated by at least accumulation.

The calibration factor is also dependent on at least one of the shape, volume, mass, or height of the neighboring particles.

Furthermore, in the present invention, said calibration factor represents the contribution of said neighboring particles to said output signal, said calibration factor being subtracted from said output signal due to said particle.

Further, in the present invention, the distances between each particle and its front neighbor particle and rear neighbor particle are respectively determined, and the distances are determined by the front neighbor particle and the rear neighbor particle. Two calibration factors, each dependent on both said output signals, are adapted to be applied to the output signal by said particles.

(Examples) Hereinafter, the present invention will be further described in detail by way of examples with reference to the accompanying drawings.

The invention relates to computing devices such as microprocessors and
Volumetric Measurement, which is the gist of each issue, No. 80/4249.
and the use of mass, volume, dimension, or shape devices of the type described in the applicant's co-pending application entitled "Grade Determination." The disclosures of both said applications are also used in this application.

In the former patent application, ore particles move through a predetermined path flanked by vertical and horizontal arrays, each consisting of a collimated and intensely pulsed light emitting diode. It is made to be. On the other hand, a vertical array corresponding to said light emitting diode array and a horizontal array of strongly collimated phototransistor light sensors are facing the light emitting diode array on the opposite side of the passage traversed by the ore particles. It is arranged like this. Each phototransistor is adapted to detect light emitted by its associated light emitting diode.

The light emitting diodes of each array are sequentially pulsed by a drive circuit and the corresponding arrays of phototransistors are scanned synchronously by scanning means. Each phototransistor is therefore responsive only to the light emitted by its corresponding light emitting diode.

As the ore particle moves past these diodes and sensors, successive areas of the ore particle extending perpendicular to the direction of movement are illuminated and scanned. Since the computer keeps track of the number of illuminated sensors, it can calculate the width of the ore particle projected in a plane transverse to the direction of movement.

The information thus obtained can be processed to obtain measurements of the length, width and height of each ore particle in three mutually orthogonal directions.
Therefore, the volume of the ore particles can be calculated.

South African Patent Application No. 80/4249 determines various criteria by which ore particles are placed into shape-related categories according to their dimensions.
The shape of ore particles is cubic, flat,
or flitch, which are defined as:

"Cube": a>b>1/2a and a>c>1/2a "flat": a>b>1/2a and c<1/2a "fritsch": b<1/2a and c<1/ 2a where a is the length, i.e. the longest linear dimension of the particle, b is the width, i.e. the longest linear dimension perpendicular to the length of the particle, and c is the height, i.e. the length of the particle. The longest linear dimension in the direction perpendicular to the length and width, respectively.

Thus, the above-mentioned application specification shows a method for calculating the volume of particles and a method for classifying each particle into categories according to the shape of the particle. Also, if a volumetric value of the particle is available, it is possible to calculate the mass of the particle by using predetermined data regarding the density of the particle.

The following disclosure provides that at least one stream of spaced apart ore particles is moved sequentially, e.g. by a conveyor belt, past a plurality of scintillation instruments, each of the scintillation The present invention relates to a radiation measuring device in which a calculator generates radiation counts due to specific particles exposed to the calculator at any given time.

Devices of this type are known in the art, and an embodiment of such a device is shown schematically in FIG. As shown in FIG. 1, the conveyor belt 10 carries a plurality of particles...P-2, P-
1, P, P+1, P+2, . . . are conveyed past a plurality of radiation detectors 12, each having its own counting area 14. The volume, mass, height or shape of each particle is determined by a measuring device 16 of the type described in South African Patent Application No. 80/4250 or South African Patent Application No. 80/4249, which Depending on the case, it is located downstream of the detector 12.

The present invention provides the following methods for counting particles (P):
by the front neighbor particle (P-1) and by the rear neighbor particle (P+
1) provides a means for correcting the contribution due to

In the present invention, when the particle (P-1) passes through the counting area of each radiation detector,
The counts from each radiation detector are summed in an accumulator 18. Particle (P-
The cumulative count for 1) is the previous neighboring particle (P
-2) and particles (P), but this component will be ignored for the time being. Particle (P-1)
Let N(P-1) represent the cumulative count for the corresponding value.

N(P-1) is stored in the memory 20 of the microprocessor device in a file assigned to particle (P-1). This memory file is denoted by M(P-1). The cumulative count N(P-1) for particle (P-1) is also used to correct the count for particle (P-2) in a manner similar to that described below.

Particle (P) follows particle (P-1) through the radiation detector and the cumulative count N(P) for particle (P) is stored in file M(P) of memory 20. Similarly, the accumulated count for particle (P+1) is stored in file M(P+1) of the microprocessor memory. Particle (P)
Front and rear neighbor particles (P-
The contributions from each of 1) and (P+1) are
The effect of the gamma radiation intensity observed by the detection varies in inverse proportion to the square of the distance from the particle to the detector, and each detection changes the solid angle subtended by the particle as seen by the radiation detector. Due to the radiation absorption effect of the shielding lead surrounding the vessel, it strongly depends on the distance between particles. The solid angle subtended by a particle as seen by a radiation detector also depends on the height or size of the particle, but for the purposes of this invention it is considered equivalent to the mass of the particle. It's fine.

The signal emitted by the particles, eg radiation emission, can be detected by a suitable detector, eg a scintillation counter. This type of detector is generally non-directional and lacks discriminatory performance. That is, if more than one particle emits radiation that reaches a detector, this type of detector cannot discern the amount of radiation actually emitted by each of these particles.

This problem is solved if these particles are spaced far enough apart that the detector corresponds to the radiation emitted by only one particle at a given time. The radiation of a particle can be monitored as it approaches and moves away from the detector. Sufficient interparticle spacing is thus required for each particle to be observed separately by the detector.

Although this arrangement provides satisfactory operation in terms of measurements, it is no longer possible to process large quantities of particles as the particle flow rate decreases as the spacing between the particles increases. The present invention is concerned with eliminating the influence on radiation counts caused by particles before and after a given particle.

The magnitude of radiation emitted from a certain particle has a characteristic that it is inversely proportional to the square of the distance from the particle. Keeping in mind that mass in this specification is equivalent to particle height or dimension, and that the particle moves horizontally over a shielded scintillation instrument, the height of the particle is It will be seen that this gives the radiation dose that reaches the scintillation instrument.

Therefore, when a particle is on a scintillation instrument, the counts recorded by the scintillation instrument are due to radiation from that particle plus radiation from other particles that reach the scintillation instrument. It is clear that the decision will be made accordingly. Particles located before and after a given particle are primarily responsible for the extra counts recorded on the scintillation instrument. The closer these front and rear particles are to the central particle, the greater the amount of radiation from these front and rear particles detected by the scintillation measuring instrument. The same applies with regard to the mass or height of the particles.

Therefore, the cumulative count N(P) can be expressed as particle (P−
1) and (P+1), the distance between particle (P) and particles (P-1) and (P+1), as well as the particle (P
-1) and (P+1) must be determined.

A particle mass determination device 16 for measuring the projected area of each particle and processing this to give the corresponding mass may be used, for example, in ”
as disclosed in the applicant's co-pending patent application entitled . This mass information for each particle is necessary to calculate the concentration or quality of the required substance contained within each particle and is utilized for the purposes of the present invention. Alternatively, device 1
6 can also easily be used to simply obtain the maximum or average height of each particle on the belt, or its shape. With this optical dimension measurement or mass measurement device, the spacing between adjacent particles can be easily obtained by methods known to those skilled in optoelectronic technology, and this information can also be used for the purpose of the present invention. be done. For example, this measuring device can measure the height of particles moment by moment, giving the linear dimensions of the particles in the direction of belt movement, and if the speed of the belt is known, the spacing between adjacent particles. It is relatively easy to obtain.
This interval measurement can be performed at a suitable reference point, for example, for the leading edge of each particle, since the leading edge of the particle can be detected if the measuring device 16 measures the height of the particle from time to time. However, it should preferably be done for the "center-to-center" spacing between adjacent particles, relative to the geometric center of the particle as determined from volumetric measurements obtained as integrals of height measurements. If the geometric center of each particle is obtained from volume measurements, it is relatively easy to calculate the spacing between particles, since the particles are traveling precisely on a belt with a known fixed velocity. be.

Particles (P-1) obtained from the volume measuring device,
The respective masses of (P) and (P+1) are the memory files M(P-1) and M of the microprocessor.
(P), M(P+1) and each interparticle spacing obtained by an optical mass measurement device or other device is also stored in the corresponding memory file M(P+1).
-1) and M(P+1).

Thus, the following information regarding particles (P-1), (P), and (P+1) will be obtained from the memory 20 of the microprocessor.

(b) the cumulative radiation count of each particle; (b) the mass of each particle, or alternatively the height, shape, or volume of each particle; and (c) the distance between adjacent particles. interval.

a matrix of calibration factors is formed from statistically measured calibration factors determined by means readily understandable to those skilled in the art;
This is permanently stored in the read-only portion 22 of the microprocessor memory.

The calibration factor is determined statistically, and
It is based on a particle's mass, volume, height, or shape, its spacing from neighboring particles, and its own cumulative radiation count.

It is clear that the calibration factor matrix contains information that depends on the height or mass of the particles, the particle spacing and the actual radiation count or level of the particles. For a given particle,
A scintillation instrument generates a radiation count that depends on the radiation level of the given particle and the radiation levels of the particles before and after it. The output of this scintillation instrument must be processed to remove unwanted levels caused by front and rear particles. For example, regarding the particle located in front, if the distance between the particle located in front and the particle being measured is known, and the height of the particle located in front is known, then A matrix of calibration factors is referenced and a factor is found that, when multiplied by the actual radiation level of the previously located particle, produces a signal to be subtracted from the output of the scintillation instrument for the particle being measured.

This description applies to each scintillation instrument and to each particle as it passes through that scintillation instrument.

It should be understood that the actual level of radiation emission is very low when dealing with small particles.
Radiation emissions also occur sparsely, so that ideally the particles must be exposed to the scintillation instrument for a fairly long period of time to obtain a true estimate of the radiation level in the particles. However, in order to obtain a satisfactory throughput, this cannot be done, nor can it be expected. For this reason, the particles pass successively through a plurality of separate scintillation instruments, and the process described above is carried out for each instrument. That is, a count signal is obtained for each particle and for each scintillation counter, and these count signals are freed from the effects of the preceding and succeeding particles, respectively. The individual count signals obtained from multiple instruments for a given particle are accumulated to obtain a more accurate estimate of the particle's radiation level.

In FIG. 2, a correction curve for particles with dimensions within a specified size range is shown as a function of shape as a function of center-to-center spacing of adjacent particles. The respective particles are, for example, described in the applicant's South African Patent Application No. 80/80 entitled "Grading", cited above
4249, on the basis of defined characteristics, such as the linear dimensions of the particles in the direction of their movement, and the direct dimensions of the particles in the vertical and horizontal directions perpendicular to the direction of movement. It can also be classified into one of several predetermined shapes. FIG. 2 shows curves for each particle having shapes designated for convenience as shapes A, B, and C, respectively.

Shape A corresponds to "fritsch", shape B to "cube", and shape C to "flat".

These curves are used as follows. For example, referring to the curve for shape A, for a center-to-center spacing of 40 mm, a particle (P) is determined by the detector it is passing through to either the front neighbor or the back neighbor, i.e. particle (P-1) or (P+1). ) can be seen to record 75% of the total radiation counts. The counting contribution by front or rear neighbor particles decreases rapidly as particle spacing increases, with particle spacing of 130 mm.
It becomes less than 10%.

Of course, the curves for shapes B and C can be used as well.

The curve in Figure 3 is similar to the above curve, but the correction factor for particles with the same mass is
It is shown as a function of particle height and center-to-center spacing. Curve A is for a 150 g spherical particle with a height of 50 mm, and curve B is for an irregular cubic particle of equal mass with a height of 25 mm. For a constant particle spacing, the "marginal effect" increases with particle height, so that the influence of front or rear neighbor particles is clearly a function of particle height.

For example, at a particle spacing of 100 mm, the A-type particles next to the front or next to the back account for 30% of the total count.
It contributes to the count of particles actually under test, but
Type B particles contribute about 22% of the total count.
It is clear that a very large number of possible correction curves must be collected in order to meet the requirements for virtually all variations in shape, size, mass, etc. of the particles to be sorted. However, statistical analysis, for example by examining a representative ore sample and also by determining the percentage of particles that have a standard preselected shape or belong to a preselected size range. It is possible to limit the number of curves by .

The percentage of count contribution from particles belonging to each of a given classification is determined by measuring the radiation counts due to each particle while varying its distance from a single detector and dividing this into a percentage of the total counts for that particle. It is determined by expressing it as . Measurements of this type can be easily made using standard laboratory techniques.

It is within the skill of one skilled in the art to accumulate this data and process it to obtain a correction curve as described above. Whether the correction factor should be determined based on particle height, mass, shape, or volume, or on other parameters.
It is determined highly empirically, based on test work with representative ores to identify the most effective correction method. The correction factor is then stored in read-only memory 22.

Then, the count correction for particles (P) is
This is done with the aid of microprocessor 24. This microprocessor is programmed appropriately by a person skilled in microprocessor programming techniques to calculate the mass of the particle (P-1) from the correction factor matrix file stored in the memory 22. , particle (P-
1) and the interval between (P) and the cumulative count of this correction factor N(P-
1) to particle (P-1) to particle (P)
We obtain a measure of the counting contribution C(P-1) made to the cumulative count N(P) of . Subtracting C(P-1) from N(P) yields the cumulative count of particle (P) without count contribution from particle (P-1). A similar correction is made to the count contribution by particle (P+1), thus obtaining a corrected count of particle (P).

FIG. 4 shows a simplified flowchart of a suitable computer program for enabling the application of correction factors. This flowchart is highly self-explanatory and shows a computational cycle for a single particle. Of course similar calculations could be done simultaneously in parallel if the detectors were in parallel rows, but it is also possible to use time-sharing techniques that allow all calculations to be done on a single processor. can. However, such considerations are unnecessary for understanding the invention.

Theoretically, in order to obtain the true counts of particles (P-1) and (P+1) to which the correction factor for particle (P) should be applied, a similar correction should also be made to these particles. Must-have. However, these are secondary corrections and can be ignored.

It should be pointed out that it is within the scope of the present invention to make multiple corrections to a given particle count. That is, since the count of a particle is significantly influenced by one or more of the shape, dimensions, i.e. volume, mass, or height, of its front or rear neighbor, corresponding multiple corrections are made to the count. It can be done.

After correcting the radiation counts as outlined above, the quality of each particle can be calculated and a decision to accept or reject it can be made by logic.

Accordingly, particle sorting can be performed by standard sorting equipment 26, such as an air blast nozzle, controlled by processor 24.

(Effect of the invention) This improvement significantly reduces the erroneous acceptance of waste rock or low-grade ore particles due to the influence of subsequent and previous neighboring particles, and the resulting reduction in the proportion of high-grade ore in the received ore. It can be improved.

[Brief explanation of the drawing]

FIG. 1 is a schematic diagram of an apparatus for carrying out the method of the invention. FIG. 2 shows a family of curves representing the correction factors for differently shaped particles as a function of interparticle spacing. FIG. 3 shows the correction curves for particles with the same mass and different heights, similar to FIG. FIG. 4 shows in simplified form a flowchart illustrating the steps used in the method and computer program of the invention. 12... Radiation detector.

Claims (1)

[Claims] 1. A method for sorting particles, comprising the steps of sequentially moving a plurality of particles past at least one detector 12 responsive to radiation emitted from the particles; and detecting the particles. generating for each particle an output signal that is dependent on the intensity of radiation emitted from the particle by a reaction in the particle, and the method includes the step of determining a spacing between neighboring particles; applying a calibration factor to the output signal by the particle of interest that depends on the spacing and on the intensity of radiation emitted from the neighboring particle; A method for sorting particles, characterized in that a method for measuring particles is used, comprising the step of providing a particle. 2. According to claim 1, the plurality of particles are sequentially moved past a plurality of detectors, and the plurality of detectors perform separate radiation counts for at least the particle of interest. A method for sorting particles, characterized in that said output signal for each particle is generated by accumulating . 3. The particle according to claim 1 or 2, characterized in that the calibration factor also depends on at least one of the shape, volume, mass, or height of the neighboring particle. Sorting method. 4. In any one of claims 1 to 3, the calibration factor represents the contribution of the adjacent particle to the output signal, and the calibration factor represents the contribution of the particle of interest to the output signal. characterized by being adapted to be subtracted from the signal;
How to sort particles. 5 In any one of claims 1 to 4, both intervals between each particle and its front neighbor particle and rear neighbor particle are respectively determined, and both the intervals and 2 depending on both the output signals by the front neighbor particle and the rear neighbor particle, respectively.
A method for selecting particles, characterized in that one calibration factor is applied to the output signal by the particle of interest.