A method for preparing a stevia residue feed

Abstract

The application discloses a preparation method of a stevia residue feed, and relates to the technical field of feed preparation, and comprises the following steps: step one, constructing an oil-in-water coarse emulsion system to make liposoluble diacetoxyscirpenes enriched on the interface of oil droplets to obtain pretreated materials; step two, performing directional biodegradation on diacetoxyscirpenes that already exist and are released through enzymolysis to make free diacetoxyscirpenes degrade to obtain detoxified materials; step three, performing anaerobic fermentation treatment, controlling the weak acidic environment of the terminal pH of fermentation to be 4.5 to obtain fermented coarse feed; and step four, detecting the estrogenic activity equivalent of the fermented coarse feed by using a biological effect test, and taking the condition that the chemical content and the estrogenic activity equivalent both meet the safety standard as the product release condition. The fresh and wet stevia residues and fresh and wet corn syrup are directly utilized, and the process of microbial enzyme synergistic fermentation can be entered without pre-drying, so that the traditional compact closed fermentation process in the prior art is shortened, and the energy consumption and carbon emission of the drying process are avoided.

Description

Technical Field

[0001] This invention relates to the field of feed preparation technology, and in particular to a method for preparing stevia residue feed. Background Technology

[0002] Stevia glycosides, a high-intensity sweetener, generate significant amounts of stevia residue (including stems, leaves, and pressing residues) during industrial extraction. Traditional treatment methods primarily involve drying followed by incineration or composting, which is not only energy-intensive and generates substantial carbon emissions but also leads to the loss of active ingredients such as polysaccharides and flavonoids, as well as fermentable carbohydrates, failing to achieve high-value feed conversion. Meanwhile, byproducts from deep corn processing, such as fresh wet corn steep liquor and corn husks, are rich in soluble proteins and carbohydrates, but commonly contain mycotoxins such as zearalenone (ZEN). ZEN exhibits estrogen-like activity, posing a serious threat to the reproductive performance of cows in late pregnancy and the peripartum period; even trace amounts can interfere with endometrial receptivity, induce early embryonic death, or cause postpartum metabolic disorders. Therefore, how to synergistically bioconvert fresh wet stevia residue and toxic corn processing byproducts to simultaneously achieve fiber degradation, protein enhancement, and toxin removal without drying has become a pressing technical problem in the field of feed resource development.

[0003] In existing technologies, there has been some exploration into the feed utilization of stevia by-products. For example, an existing technology discloses a microbial-enzyme co-fermented feed containing stevia residue. This feed uses stevia residue pre-fermented with Aspergillus niger, combined with conventional concentrates such as corn and soybean meal, and fermented in a breathing bag at 25-30°C for 5-7 days with a compound microbial agent (Lactobacillus casei, Lactobacillus plantarum, Candida utilis, and Saccharomyces cerevisiae) to prepare a general-purpose fermented feed. Although this technology can improve the palatability and nutritional value of stevia residue to some extent, it requires the raw materials to be pre-fermented with Aspergillus niger and dried, and does not involve the direct utilization of fresh wet raw materials. More importantly, this technology does not address the issue of bound mycotoxins that may exist in the raw materials, nor does it establish a quality control system for the bioavailability of estrogen-active substances. Its application scenarios are laying hens, fattening pigs, and dairy cows during the lactation period, and it has not been adapted for the physiological stage of late pregnancy and peripartum cows, which are extremely sensitive to ZEN.

[0004] Another technology discloses a method for preparing mixed roughage for beef cattle through microbial fermentation of stevia by-products. This method involves mixing fresh stevia by-products with feed such as mulberry and soybean residue, then treating the mixture stepwise with cellulase, mannanase, and multiple microbial strains (Lactobacillus plantarum, Bacillus licheniformis, Saccharomyces cerevisiae, etc.), followed by compacted and sealed fermentation for 40-45 days. While this technology achieves the utilization of fresh stevia by-products without drying, its lengthy fermentation cycle and the resulting highly acidic environment (pH around 4.3) pose potential adverse effects on the rumen teats of pregnant cows in late pregnancy and on postpartum feed intake recovery. Furthermore, this technology only suppresses contaminating bacteria through conventional silage methods, without targeted enzymatic release and biodegradation of the free and bound ZEN already present in the raw materials. It also fails to introduce estrogen activity equivalent (EEC) testing as a safety release standard, thus failing to meet the rigid requirement of zero tolerance for ZEN in breeding cows. Summary of the Invention

[0005] The purpose of this invention is to provide a method for preparing stevia residue feed to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a method for preparing stevia residue feed, comprising the following steps: Step 1: The fresh wet stevia residue after stevia glycoside extraction is subjected to differential particle size pulverization to form coarse, medium and fine powder segments. The coarse, medium and fine powder segments are pretreated with differential moisture content. The fine powder segment is then mixed with fresh wet corn steep liquor containing bound and free gibberellic acid. The bound gibberellic acid is released by enzymatic hydrolysis. The mixture is then mixed with the coarse and medium powder segments and tomato cake meal as an oily protein adjuvant. The mass ratio of stevia residue, tomato cake meal and corn steep liquor is controlled at (75~80):(15~20):(3~5). The oil content of the mixture is controlled at 4~8%. A crude emulsion system of oil in water is constructed so that the fat-soluble gibberellic acid is enriched at the oil droplet interface to obtain the pretreated material. Step 2: Inoculate the pretreated material obtained in Step 1 with enzyme-producing microorganisms containing Aspergillus niger for aerobic pre-detoxification treatment. This process involves the targeted biodegradation of existing and enzymatically released juvenile ketones, resulting in the degradation of free juvenile ketones and the acquisition of detoxified material. Step 3: Inoculate the detoxified material obtained in Step 2 with compound microbial agent and carry out anaerobic fermentation treatment. Control the pH at the end of fermentation to a weakly acidic environment of 4.8~5.2 to obtain fermented roughage. Step 4: Detect the chemical content of zearalenone, α-zearalenone alcohol, zearalenone and their conjugated compounds in fermented roughage, and use biological effect tests to detect their estrogen activity equivalent. The product is released only if both the chemical content and the estrogen activity equivalent meet the safety standards. Stevia residue feed is suitable for pregnant cows in late pregnancy and peripartum period who have zero tolerance for juvenile ketone.

[0007] Preferably, in step one, the coarse particle size is 10-15mm, the medium particle size is 6-8mm, and the fine particle size is ≤2mm. The mass ratio of the three is (20-30):(50-60):(10-20), and the sum of the mass parts of the three is 100 parts. The moisture content of the coarse powder segment is 60-70%, the moisture content of the medium powder segment is 60-65%, and the moisture content of the fine powder segment is 55-60%. The moisture content of fresh wet stevia residue is 60-70%, and the moisture content of fresh wet corn liquor is 45-55%.

[0008] Preferably, in step one, the stevia residue includes coarse segments with a particle size of 10-15 mm. After differential moisture content pretreatment, the moisture content of the coarse segments is 60-68%. Then, it is pretreated by atomized spraying with a calcium bicarbonate saturated solution to control the weight gain of the coarse segments at 3-5%. The coarse segments are then left to stand at 25-30°C for 2-4 hours to form a microporous etching structure with a depth of 50-200 μm on the surface of the coarse segments. The moisture content of the coarse segments after spraying is controlled to not exceed 70%. The oil content of tomato cake meal is 25-30%, which forms an oil-in-water crude emulsion system with the fibrous phase of stevia residue and the aqueous phase of corn syrup.

[0009] Preferably, the enzyme-producing microorganism in step two is a compound inoculum of Aspergillus niger and Bacillus subtilis, wherein the viable Aspergillus niger count is 1×10⁻⁶. 8 ~5×10 8 CFU / g dry weight of material, the number of viable Bacillus subtilis is 20 to 60 times that of viable Aspergillus niger; the aerobic pre-detoxification treatment opens the benzodifuran ring of jugazocarpine through extracellular laccase and peroxidase secreted by Aspergillus niger, and breaks its lactone bond through lactone hydrolase of Bacillus subtilis, so that jugazocarpine is degraded into metabolites without estrogen activity. The degradation targets are the free gibberellenone already present in the material and the bound gibberellenone released by enzymatic decomposition in step one. The conditions for aerobic pre-detoxification treatment are: ventilation rate 0.5~1.0 vvm, temperature 32~35℃, and reaction time 2.0~2.5h; the viable bacteria count is based on the dry weight of the pretreated material obtained in step one.

[0010] Preferably, the aeration rate in step two is dynamically positively correlated with the amount of corn steep liquor added in step one. When the mass percentage of corn steep liquor is 3-5%, the following condition is met: The ventilation rate (vvm) = 0.5 + 0.1 × (corn steep liquor mass percentage - 3%), and the ventilation rate does not exceed 1.0vvm.

[0011] Preferably, the compound microbial agent in step three consists of *Lactobacillus plantarum* and *Saccharomyces cerevisiae*, wherein the viable count of *Lactobacillus plantarum* is 7 × 10⁻⁶. 9 ~9×109 The CFU / g dry weight of the material and the viable count of brewer's yeast were 3×10¹¹~4.5×10¹¹ CFU / g dry weight of the material, with a viable count ratio of 1:(30~70); the anaerobic fermentation treatment was carried out at 28~32℃ for 5~7 days; in step three, 0.2~0.5% stevia polyphenols and 1.5~2.5% calcium bicarbonate were also added; the viable counts were all based on the dry weight of the pretreated material obtained in step one.

[0012] Preferably, at the end of step two, the content of free gibberellic acid is detected and controlled to be ≤50μg / kg; At the end of step three, the content of extractable jugaprone was detected and controlled to be ≤10μg / kg; In step four, when the product is released, the sum of the contents of α-zearalenone and zearalenone is tested and controlled to be ≤5μg / kg, and free zearalenone is not detected.

[0013] Preferably, in step four, a biological effect assay is used to detect the estrogen activity equivalent, specifically as follows: The estrogenic activity equivalent of the fermented roughage extract was determined by a recombinant yeast estrogen screening test, and the estrogenic activity equivalent ≤1μgZENeq / kg was used as an additional release standard. The determination of estrogen activity equivalent includes a pre-hydrolysis step, in which the extract is pretreated with β-glucuronidase and sulfatase to release the conjugated estrogen active substances before the total estrogen activity equivalent is determined. The conjugated states include gibbenone-14-β-D-glucoside and gibbenone sulfate; If the chemical test is qualified but the estrogen activity equivalent exceeds the standard, 0.1-0.2% stevia polyphenols should be added to the batch of fermented roughage, and anaerobic ripening treatment should be continued for 2-3 days under sealed conditions until the estrogen activity equivalent reaches the standard.

[0014] Preferably, the initial pH value is the pH of the material after the aerobic pre-detoxification in step two, which is 5.8~6.5. The anaerobic fermentation treatment in step three is divided into two stages: The first stage is a rapid acid reduction period of 0~48h, during which the pH is controlled to decrease from the initial value to 5.0~5.2, and the pH decrease rate is 0.3~0.5 / day; The second stage is the stable fermentation period after 48 hours, during which the pH fluctuation range is controlled to be ≤0.2, and the pH is maintained at 4.8~5.0; The final fermentation pH of 4.8-5.2 is higher than that of conventional silage (pH 4.0-4.5). The weak acid preservation is achieved through the synergistic antibacterial effect of stevia polyphenols and lycopene. The molar composition of volatile fatty acids in fermented roughage is acetic acid: propionic acid: butyric acid = (6-8): (2-3): (0.5-1), and the content of γ-aminobutyric acid is ≥30mg / kg.

[0015] The preferred method for supplementing stevia polyphenols is as follows: Stevia polyphenols were dissolved in a weakly alkaline buffer solution with a pH of 7.0-7.5 to prepare a 10% polyphenol solution. The solution was then evenly applied to the surface of fermented roughage using a mist spraying method, with the weight gain rate controlled at 1-2%. Anaerobic post-ripening treatment under closed conditions controls the oxidation-reduction potential (ORP) to be ≤-200mV, and the residual rate of stevia polyphenols in fermented roughage after post-ripening is ≥60%.

[0016] The technical effects and advantages of this invention are as follows: (1) This invention directly utilizes fresh wet stevia residue and fresh wet corn liquor with a moisture content of 60%~70%, which can be put into the bacterial-enzyme co-fermentation process without pre-drying. The fermentation cycle is only 5~7 days, which is shorter than the traditional compacted closed fermentation of 40~45 days in the prior art, and avoids the energy consumption and carbon emissions of the drying process. At the same time, the fresh wet state retains the heat-sensitive active ingredients such as natural polyphenols and flavonoids in the stevia residue, which can play an antioxidant and synergistic antibacterial function in the subsequent fermentation. This solves the technical bottleneck of the prior art that fresh wet raw materials must be dried before processing and that the long fermentation cycle leads to the loss of active ingredients. (2) In step one, the present invention pretreats the fine powder-corn steep liquor mixture with β-glucosidase and sulfatase to fully release the cryptic ZEN in glucosidase-bound and sulfatase-bound states; then in step two, the benzodifuran ring of ZEN is opened by Aspergillus niger extracellular laccase / peroxidase, and its lactone bond is broken by Bacillus subtilis lactone hydrolase, thereby directionally degrading ZEN into metabolites without estrogen activity, thus achieving effective eradication of cryptic ZEN in fresh wet plant substrates and filling the technical gap in the release and degradation of bound toxins. (3) In step four, this invention not only detects chemical indicators such as free ZEN, α-ZOL, and ZAN, but also introduces a recombinant yeast estrogen screening test (YES) and a pre-hydrolysis step (β-glucuronidase + sulfatase), so that the conjugated bound state and unknown estrogen active substances can be fully identified at the biological level. This invention uses a weak alkaline buffer solution to atomize and spray polyphenols + ORP≤-200mV for in-situ treatment with closed post-ripening. It can reduce EEC to 0.4μgZENeq / kg within 2-3 days without returning the material to the main fermenter, which not only ensures batch isolation management for continuous production, but also avoids the aseptic return problem of traditional return-to-tank processes. (4) This invention breaks through the technical prejudice that silage must have a pH ≤ 4.5 to maintain its quality for a long time. Through staged anaerobic fermentation and a double buffer system of stevia polyphenols / calcium bicarbonate, the pH at the end of fermentation is stabilized in the weakly acidic range of 4.8~5.2. The weakly acidic environment inhibits the growth of mold (shelf life ≥ 6 months) and avoids damage to the rumen teats of cows caused by excessive acidification. At the same time, the high proportion of acetic acid is suitable for the supply of precursors for mammary gland fatty acid synthesis, while GABA effectively alleviates peripartum stress and reduces the incidence of postpartum ketosis. Attached Figure Description

[0017] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention, but do not constitute a limitation thereof. In the drawings: Figure 1 This is a schematic diagram of the preparation method of the present invention. Detailed Implementation

[0018] The following will refer to the appendices in the embodiments of the present invention. Figure 1 The technical solutions in the embodiments of the present invention are clearly and completely described herein. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0019] Example 1: A method for preparing stevia residue feed, comprising the following steps: Step 1: Take 80 kg of fresh, wet stevia residue (65% moisture content) after stevia glycoside extraction. After differential particle size reduction, divide it into three parts: 20 parts coarse (12 mm), 60 parts medium (7 mm), and 20 parts fine (≤2 mm), totaling 100 parts by mass. Pre-treat each part with a different moisture content: 65% for coarse, 63% for medium, and 58% for fine. Mix the fine part with 5 kg of fresh, wet corn steep liquor (50% moisture content, including glucosinolate and sulfate-bound forms of gibberellic acid, total ZEN 200 μg / kg) containing both bound and free gibberellic acid. Add 80 U / kg of β-glucosidase (based on dry weight) and 30 U / kg of sulfatase (based on dry weight), and enzymatically hydrolyze at 33°C for 0.8 h to release the bound gibberellic acid. Then mix with the coarse segment, the middle segment and 15 kg of tomato cake meal as an oily protein auxiliary material, and control the mass ratio of stevia residue, tomato cake meal and corn steep liquor to 80:15:5 to construct an oil-in-water crude emulsion system, so that the fat-soluble gibberellic acid is enriched at the oil droplet interface to obtain the pretreated material. Step 2: Inoculate the pretreated material with a compound inoculum of Aspergillus niger and Bacillus subtilis, wherein the viable Aspergillus niger count is 3 × 10⁻⁶.8 The CFU / g dry weight of the material showed that the viable count of Bacillus subtilis was 30 times that of Aspergillus niger. Aerobic pre-detoxification treatment involved the opening of the benzodifuran ring of jugazocarpine by extracellular laccase and peroxidase secreted by Aspergillus niger, and the cleavage of its lactone bond by lactone hydrolase of Bacillus subtilis, thus degrading jugazocarpine into metabolites without estrogenic activity. The aeration rate was controlled at 0.7 vvm (according to the formula: corn steep liquor mass percentage 5%, aeration rate (vvm) = 0.5 + 0.1 × (5-3) = 0.7), the temperature was 34℃, and the aerobic treatment lasted for 2.2 h, degrading the free jugazocarpine and obtaining the detoxified material. Step 3: Inoculate the detoxification material with a compound inoculum of Lactobacillus plantarum and Saccharomyces cerevisiae, wherein the viable count of Lactobacillus plantarum is 8 × 10⁻⁶. 9 The CFU / g dry weight of the material and the viable count of Saccharomyces cerevisiae were 3.5 × 10¹¹ CFU / g dry weight of the material, with a viable count ratio of 1:43.75. Stevia polyphenols (0.3%) and calcium bicarbonate (2.0%) were added, and anaerobic fermentation was carried out at 30℃ for 6 days. The anaerobic fermentation treatment was divided into two stages: the first stage was a rapid acid reduction period from 0 to 48 hours, controlling the pH to decrease from the initial value of 6.2 to 5.1 at a rate of 0.275 / day; the second stage was a stable fermentation period after 48 hours, controlling the pH fluctuation range to ≤0.15 and maintaining the pH at 4.85~5.0. The final fermentation pH of 4.9 was higher than the pH of conventional silage (4.0~4.5), achieving weak acid preservation through the synergistic antibacterial effect of stevia polyphenols and lycopene. The molar composition of volatile fatty acids in the fermented roughage was acetic acid:propionic acid:butyric acid = 7.0:2.4:0.6, and the γ-aminobutyric acid content was 38 mg / kg. Step 4: The chemical content of jujubene, α-zearalenone, zearalenone, and their conjugated compounds (including jujubene-14-β-D-glucoside and jujubene sulfate) in the fermented roughage was determined, and the estrogenic activity equivalent was detected using the recombinant yeast estrogen screening test (YES). The estrogenic activity equivalent determination included a pre-hydrolysis step, in which the extract was pretreated with β-glucuronidase and sulfatase to release the conjugated estrogenic active substances before the total estrogenic activity equivalent was measured.

[0020] At the end of step two, the free zearalenone content was 28 μg / kg; at the end of step three, the extractable zearalenone content was 6 μg / kg; when the product was released in step four, the sum of α-zearalenol and zearalenone content was 2.8 μg / kg, and free zearalenone was not detected; the estrogen activity equivalent was 0.5 μg ZENeq / kg, and both chemical testing and biological potency were qualified, requiring no post-ripening treatment.

[0021] Overall indicators: crude protein increased by 28% compared to raw materials, acid-soluble protein content was 16.5%, fiber degradation rate was 52%, and total ZEN degradation rate was 97.2%.

[0022] Example 2: The difference between this example and Example 1 is that the coarse section in step one is not pretreated by atomized spraying with a saturated calcium bicarbonate solution, i.e., the microporous etching structure construction step is omitted, and the coarse section is directly mixed with other materials. The remaining steps, parameters, and amount of bacterial agent added are the same as in Example 1.

[0023] At the end of step two, the content of free zearalenone was 35 μg / kg; at the end of step three, the content of extractable zearalenone was 14 μg / kg; when the product was released in step four, the sum of the contents of α-zearalenol and zearalenone was 4.2 μg / kg, and free zearalenone was not detected; the estrogen activity equivalent was 1.2 μg ZENeq / kg.

[0024] Overall indicators: crude protein increased by 26%, acid-soluble protein by 15.2%, fiber degradation rate by 48%, and total ZEN degradation rate by 91.5%. Compared with Example 1, because the waxy layer on the surface of the coarse segment was not micro-etched, Aspergillus niger hyphae had difficulty penetrating into the interior of the coarse segment, resulting in a significant increase in ZEN residue inside the coarse segment and a decrease in the total degradation rate by 5.7 percentage points.

[0025] Example 3: The difference between this example and Example 1 is that step three does not involve staged pH control; instead, it involves anaerobic fermentation at a constant temperature of 30°C for 6 days, allowing the pH to naturally decrease to 4.3. The rate of pH decrease and the fluctuation range during the stabilization period are not controlled. All other steps and parameters are the same as in Example 1.

[0026] At the end of step two, the free zearalenone content was 32 μg / kg; at the end of step three, the extractable zearalenone content was 9 μg / kg; when the product was released in step four, the sum of the α-zearalenol and zearalenone content was 3.5 μg / kg; and the estrogen activity equivalent was 0.8 μg ZENeq / kg.

[0027] Overall indicators: Final pH 4.3; the molar composition of volatile fatty acids was acetic acid:propionic acid:butyric acid = 4.8:2.3:0.7, with butyric acid increasing to 9%, producing a noticeable rancid odor; γ-aminobutyric acid content decreased to 22 mg / kg. Compared with Example 1, the lack of phased control led to excessively rapid acid reduction in the early stage and pH fluctuations of ±0.45 in the later stage, causing the composition of anaerobic fermentation products to deviate from the range suitable for the rumen of cows.

[0028] Example 4: The difference between this example and Example 1 is that the dynamic ventilation formula is not used in step two; instead, the ventilation rate is fixed at 0.5 vvm. The remaining steps and parameters are the same as in Example 1.

[0029] At the end of step two, the free zearalenone content was 48 μg / kg; at the end of step three, the extractable zearalenone content was 18 μg / kg; when the product was released in step four, the sum of the α-zearalenol and zearalenone content was 6.5 μg / kg; and the estrogen activity equivalent was 2.1 μg ZENeq / kg.

[0030] Overall index: Total ZEN degradation rate 88.3%. Compared with Example 1, due to insufficient aeration when the corn steep liquor addition was 5%, the dissolved oxygen supply in the aerobic stage was insufficient, and the synergistic detoxification efficiency of Aspergillus niger and Bacillus subtilis decreased significantly, resulting in incomplete degradation of free ZEN in step two and increased cumulative degradation pressure in the subsequent anaerobic stage.

[0031] Example 5: The difference between this example and Example 1 is that stevia polyphenols and calcium bicarbonate are not added in step three, while the remaining microbial agents and fermentation conditions are the same as in Example 1.

[0032] At the end of step two, the free zearalenone content was 38 μg / kg; at the end of step three, the extractable zearalenone content was 12 μg / kg; when the product was released in step four, the sum of the α-zearalenol and zearalenone content was 4.8 μg / kg; and the estrogen activity equivalent was 1.5 μg ZENeq / kg.

[0033] Overall indicators: final pH 4.6, pH fluctuation range during fermentation reached ±0.25; γ-aminobutyric acid content 28 mg / kg; acid-soluble protein 14.0%, fiber degradation rate 42%. Compared with Example 1, the lack of stevia polyphenols for antioxidant synergistic antibacterial effect and calcium bicarbonate for pH buffering led to decreased fermentation stability, and both crude protein conversion rate and fiber degradation rate were significantly reduced.

[0034] Example 6: The preparation process of this example is basically the same as that of Example 1, but a boundary situation occurred in the detection of step four: the chemical test was qualified (the sum of α-zearalenone and zearalenone content was 4.5 μg / kg, and free zearalenone was not detected), but the estrogen activity equivalent value was 1.3 μg ZENeq / kg (>1 μg ZENeq / kg), which triggered the post-ripening treatment.

[0035] Post-ripening treatment: Stevia polyphenols were added to this batch of fermented roughage. The addition method was as follows: Stevia polyphenols were dissolved in a pH 7.2 disodium hydrogen phosphate-sodium dihydrogen phosphate buffer solution to prepare a 10% (w / w) polyphenol solution. This solution was then evenly applied to the surface of the fermented roughage using a mist spray method, controlling the spray weight gain rate at 1.5%. The material was transferred to an independent post-ripening tank, and nitrogen gas was introduced to purge until the oxidation-reduction potential (ORP) reached -220 mV (≤ -200 mV). Anaerobic post-ripening was continued for 2.5 days under sealed conditions at 30°C.

[0036] After post-ripening, the estrogen activity equivalent decreased to 0.4 μg ZENeq / kg; the stevia polyphenol residue rate in the fermented roughage was 65%; and other chemical indicators remained unchanged. This demonstrates that in-situ post-ripening treatment can achieve the target biopotency through targeted delivery of polyphenols and a strongly reducing environment without returning the feed to the fermentation tank.

[0037] Example 7: The difference between this example and Example 1 is that: in step one, the coarse section has a water content of 68% after differential water content pretreatment, and then is pretreated by atomized spraying of calcium bicarbonate saturated solution to control the weight gain of the coarse section to 5%, and is left to stand at 30°C for 4 hours to form a microporous etching structure with a depth of 180μm on the surface of the coarse section, and the water content of the coarse section after spraying is 69.0%.

[0038] At the end of step two, the free zearalenone content was 30 μg / kg; at the end of step three, the extractable zearalenone content was 7 μg / kg; when the product was released in step four, the sum of the α-zearalenol and zearalenone content was 3.0 μg / kg; and the estrogen activity equivalent was 0.6 μg ZENeq / kg.

[0039] Overall performance: Total ZEN degradation rate was 97.0%, which is basically the same as in Example 1.

[0040] Example 8: The difference between this example and Example 1 is that the number of viable Aspergillus niger bacteria in step two is 1×10⁻⁶. 8 CFU / g dry weight of material; Bacillus subtilis viable count is 20 times that of Aspergillus niger viable count; Lactobacillus plantarum viable count in step three is 7×10 9 The number of live bacteria in the brewing yeast was 3×10¹¹ CFU / g dry weight of the material, and the ratio of live bacteria in the two was approximately 1:43.

[0041] At the end of step two, the free zearalenone content was 38 μg / kg; at the end of step three, the extractable zearalenone content was 9 μg / kg; when the product was released in step four, the sum of the α-zearalenol and zearalenone content was 4.5 μg / kg; and the estrogen activity equivalent was 0.9 μg ZENeq / kg.

[0042] Overall performance: Total ZEN degradation rate 95.5%. This demonstrates that even under the condition of the lower limit of viable bacterial count, the system can still meet the quality control standards, but the detoxification efficiency is close to the critical value.

[0043] Example 9: The difference between this example and Example 1 is that the number of viable Aspergillus niger bacteria in step two is 5 × 10⁻⁶. 8 CFU / g dry weight of material; Bacillus subtilis viable count is 60 times that of Aspergillus niger viable count; Lactobacillus plantarum viable count in step three is 7×10 9The CFU / g dry weight of the material and the viable count of brewer's yeast is 4.5×10¹¹CFU / g dry weight of the material.

[0044] At the end of step two, the free zearalenone content was 15 μg / kg; at the end of step three, the extractable zearalenone content was 3 μg / kg; when the product was released in step four, the sum of the α-zearalenol and zearalenone content was 1.8 μg / kg; and the estrogen activity equivalent was 0.3 μg ZENeq / kg.

[0045] Overall performance: Total ZEN degradation rate of 98.5%. This demonstrates that under the condition of the upper limit of the proportion of viable bacteria in the strain, the detoxification efficiency is significantly improved and the system redundancy is sufficient.

[0046] Example 10: The difference between this example and Example 1 is that the mass percentage of corn syrup in step one is 3%, and the aeration rate in step two is calculated according to the formula: aeration rate (vvm) = 0.5 + 0.1 × (3-3) = 0.5vvm.

[0047] At the end of step two, the free zearalenone content was 42 μg / kg; at the end of step three, the extractable zearalenone content was 8 μg / kg; when the product was released in step four, the sum of the α-zearalenol and zearalenone content was 3.8 μg / kg; and the estrogen activity equivalent was 0.7 μg ZENeq / kg.

[0048] Overall performance: Total ZEN degradation rate 96.8%. This demonstrates that even under low corn steep liquor loading conditions, the lower limit of the aeration rate formula can still meet the aerobic detoxification requirements.

[0049] Example 11: The difference between this example and Example 1 is that: to verify the boundary condition that the oil content of the mixture is close to the lower limit (4%), tomato cake meal raw material with an oil content of 20% is used and blended in a ratio of 80:15:5.

[0050] At the end of step two, the free zearalenone content was 36 μg / kg; at the end of step three, the extractable zearalenone content was 10 μg / kg; when the product was released in step four, the sum of the α-zearalenol and zearalenone content was 4.0 μg / kg; and the estrogen activity equivalent was 0.9 μg ZENeq / kg.

[0051] Overall index: Total ZEN degradation rate 95.8%. Compared with Example 1, due to the lower limit of oil content in tomato cake meal, the interfacial enrichment factor of fat-soluble gibberellic acid in the oil-in-water crude emulsion system decreased to 1.8 times, but effective targeted biodegradation can still be maintained.

[0052] Example 12: The difference between this example and Example 1 is that in step three, the pH decrease rate in the first stage is controlled at 0.3 / day, taking approximately 3.3 days to decrease from the initial value of 6.2 to 5.2; in the second stage, the pH is maintained at 4.8~5.0, with a fluctuation range ≤0.2. The molar composition of volatile fatty acids is acetic acid:propionic acid:butyric acid = 6.0:2.0:0.5.

[0053] At the end of step two, the free zearalenone content was 33 μg / kg; at the end of step three, the extractable zearalenone content was 8 μg / kg; when the product was released in step four, the sum of the α-zearalenol and zearalenone content was 3.5 μg / kg; and the estrogen activity equivalent was 0.8 μg ZENeq / kg.

[0054] Overall indicators: γ-aminobutyric acid (GABA) content was 32 mg / kg, and total ZEN degradation rate was 96.5%. This demonstrates that the system can still meet the rumen compatibility requirements of cows under the lower limits of the staged fermentation parameters.

[0055] Comparative Example 1: 60 kg of fresh stevia by-products, 40 kg of mulberry feed, and 15 kg of soybean residue were taken. The stevia by-products were sieved through a 6-8 mm sieve without distinguishing between coarse / medium / fine powder segments, without differential moisture content pretreatment, without micro-etching, and without decomposition and release of bound Zn enzymes. 3 × 10⁻⁶ cellulase was added. 4 U / kg, mannanase 4×10³U / kg, lactic acid bacteria 4×10 7 cfu / kg, Bacillus subtilis 2×10¹ 0 CFU / kg, incubated at 35℃ for 1.3 hours; then mixed with mulberry and soybean residue feed, with 0.6% earthworm selenium added, compacted and sealed for fermentation for 44 days.

[0056] At the end of step two, the free zearalenone content was 85 μg / kg; at the end of fermentation, the extractable zearalenone content was 42 μg / kg; the sum of α-zearalenol and zearalenone content was 15.6 μg / kg; the conjugated zearalenone residue was 8.2 μg / kg; and the estrogen activity equivalent was 4.8 μg ZENeq / kg.

[0057] Overall indicators: total ZEN degradation rate 68.5%, final pH 4.3, crude protein increase 18%, acid-soluble protein 11.2%, fiber degradation rate 35%. Compared with Example 1, this traditional method has a longer fermentation cycle, does not target the enzymatic release of bound ZEN, and the final pH is strongly acidic and unsuitable for the rumen physiology of cows.

[0058] Comparative Example 2: 100 kg of mixed corn and barley feed was screened and dried. Bacillus subtilis powder was dissolved in distilled water at a ratio of 1:100 to obtain a bacterial solution, which was then mixed with the feed ingredients. Chopped, moist hay was alternately laid between the feed ingredients, and the solution was sprayed onto the hay by soaking it in the bacterial solution, maintaining the pH of the pre-fermented material at 8.5-9.0. The mixture was then sealed and stored. Subsequently, a mixed yeast solution (Anomala spp., Mickey Mouse yeast, Kluyveromyces lactis, and lactic acid bacteria) was sprayed onto the surface of the pre-fermented material, and anaerobic fermentation was carried out for 7 days, resulting in a final pH of 3.8.

[0059] Because the raw material was dried grains rather than fresh stevia residue, and the technical objective was to inhibit the growth of toxin-producing bacteria rather than degrade existing toxins, almost no bound zineb (including glucoside-bound and sulfate-bound forms) was treated in the raw material. After fermentation, the bound ZEN residue was 28.5 μg / kg, the free ZEN residue was 55 μg / kg, the total ZEN degradation rate was 52.0%, and the estrogen activity equivalent was 5.2 μgZENeq / kg.

[0060] Overall indicators: final pH 3.8, crude protein increased by 20%, acid-soluble protein 12.5%. Compared with Example 1, this scheme does not involve the bound ZEN enzyme decomposition and release, oil-in-water crude emulsion system and directional biodegradation mechanism of this application, and the drying process has high energy consumption, making it unsuitable for the direct utilization of fresh wet stevia residue.

[0061] Comparative Example 3: The difference between this comparative example and Example 1 is that in step one, after the fine powder is mixed with fresh wet corn steep liquor, it is not treated with β-glucosidase and sulfatase, that is, the bound juvenile ketone is not released. The fine powder-corn steep liquor mixture is directly mixed with the coarse powder, medium powder, and tomato cake meal. The remaining steps, inoculants, and fermentation conditions are the same as in Example 1.

[0062] At the end of step two, the free zearalenone content was 30 μg / kg; at the end of step three, the extractable zearalenone content was 22 μg / kg; when the product was released in step four, the sum of α-zearalenol and zearalenone content was 5.5 μg / kg; the conjugated zearalenone residue was as high as 35.8 μg / kg; and the estrogen activity equivalent was 3.1 μg ZENeq / kg.

[0063] Overall index: Total ZEN degradation rate 78.6%. Compared with Example 1, due to the omission of the enzymatic hydrolysis step, the glucosinolate-bound and sulfate-bound gibbenone in corn steep liquor and fine powder were not released, resulting in the inability to access this part of the toxin in subsequent aerobic and anaerobic stages. Finally, the conjugated bound state residue was the highest value in the entire table, proving that the enzymatic hydrolysis in step one to release bound ZEN is a necessary pre-step in the entire detoxification chain.

[0064] Comparative Example 4: The difference between this comparative example and Example 1 is as follows: In Step 3, no staged pH control is performed, and no steviol polyphenols and calcium bicarbonate are added, allowing the material to ferment naturally until the pH reaches 4.2; in Step 4, only chemical content detection is performed, and no recombinant yeast estrogen screening test (YES) and pre-hydrolysis step are carried out, that is, the estrogen activity equivalent is not detected.

[0065] At the end of Step 2, the content of free zearalenone was 29 μg / kg; at the end of Step 3, the content of extractable zearalenone was 7 μg / kg; during the chemical detection in Step 4, the sum of the contents of α-zearalenol and zearalanone was 3.0 μg / kg, and free zearalenone was not detected. It was judged as qualified according to the chemical indicators.

[0066] Supplementary detection: The EEC detection was performed on this batch of samples as a supplement, and the result was 1.1 μg ZENeq / kg. The final pH was 4.2, and the molar composition of volatile fatty acids was acetic acid:propionic acid:butyric acid = 5.0:2.5:0.8, with the proportion of butyric acid being 10%, and the GABA content was 30 mg / kg.

[0067] Verification by feeding cows: This batch of fermented roughage was fed to late-pregnant Holstein cows (n = 10) at a proportion of 40% of the dry matter of the diet, and the rumen pH was continuously monitored. The results showed that the daily fluctuation range of the rumen pH was ±0.35, which was significantly worse than ±0.12 in Example 1.

[0068] Conclusion: This comparative example proves that even if the chemical detection is qualified, due to the lack of weak acid quality preservation design and EEC biological potency quality control in traditional strong acid fermentation, there are risks of poor rumen physiological adaptability and potential missed detection of biological potency, and it cannot be directly applied to late-pregnant and peripartum cows with zero tolerance for zearalenone.

[0069] Table 1 Comparison table of raw material pretreatment and process parameters

[0070] As shown in Table 1, Example 1, employing a combination of fresh wet raw materials, particle size differentiation, moisture content differentiation, micro-etching, enzyme decomposition and release, dynamic aeration, and staged anaerobic fermentation, achieved a fermentation cycle of only 6 days and a final pH stable at a weakly acidic environment of 4.9, with process parameters at optimal equilibrium. Example 2 omitted the micro-etching treatment, Example 3 omitted the staged fermentation control (final pH reduced to 4.3), Example 4 used a fixed aeration rate (0.5 vvm), and Example 5 omitted stevia polyphenols and calcium bicarbonate additives (final pH 4.6), all of which deviated from the optimal combination of process parameters. Example 6 underwent post-ripening treatment due to triggering EEC exceedance, extending the total cycle to 8.5 days. Examples 7-12 verified the boundaries of moisture content, strain ratio, aeration rate, oil content, and staged parameter boundaries, respectively. Comparative Example 1 employed traditional long-cycle compacted closed fermentation (44 days, final pH 4.3); Comparative Example 2 used dried raw materials and a detoxification approach (final pH 3.8); Comparative Example 3 omitted the enzymatic hydrolysis step; and Comparative Example 4 employed traditional strong acid fermentation (final pH 4.2). The process parameters of these examples differ fundamentally from those of this application. This demonstrates that differences in raw material state, particle size, micro-etching, enzymatic decomposition and release, dynamic control of aeration rate, and staged fermentation conditions all significantly affect the controllability of the fermentation process and the quality of the final product.

[0071] Table 2 Comparison of aerobic-anaerobic fermentation process parameters and final product pH

[0072] As shown in Table 2, Example 1 used a complete combination of Aspergillus niger to Bacillus subtilis ratio of 1:30, dynamic aeration rate of 0.7 vvm, Lactobacillus plantarum to Saccharomyces cerevisiae ratio of 1:43.75, and added stevia polyphenols and calcium bicarbonate. The fermentation cycle was 6 days, with a final pH of 4.9, and the aerobic-anaerobic transition was optimal. Example 4 used a fixed aeration rate of 0.5 vvm, which was 28.6% lower than the 0.7 vvm in Example 1, resulting in insufficient dissolved oxygen supply during the aerobic stage. Example 5 removed the additives, with a final pH of 4.6, and the fermentation stability decreased. Example 8 reduced the ratio of Aspergillus niger to Bacillus subtilis to 1:20 (lower limit), while Example 9 increased it to 1:60 (upper limit). Comparative Example 1 used a conventional combination of lactic acid bacteria and Bacillus subtilis without aeration control, resulting in a fermentation cycle of up to 44 days and a final pH of 4.3. Comparative Example 2 used only Bacillus subtilis and anomalous Wickham yeast, without dynamic aeration design, and achieved a final pH of 3.8. Comparative Example 4 removed the additives, resulting in a final pH of 4.2. This demonstrates that the specific ratio of aerobic strains, the dynamic correlation between aeration and corn steep liquor load, and the synergistic buffering effect of additives during the anaerobic stage are key to shortening the fermentation cycle and maintaining a slightly acidic preservation environment.

[0073] Table 3 Comparison of Toxin Degradation Spectrum and Biosafety

[0074] As shown in Table 3, Example 1 exhibited the best toxin degradation indicators: free ZEN decreased to 28 μg / kg in step two, and extractable ZEN decreased to 6 μg / kg in step three. The final product contained only 2.8 μg / kg of α-ZOL+ZAN, with no detectable conjugated ZEN. The total ZEN degradation rate was 97.2%, and the EEC was 0.5 μgZENeq / kg, far below the release standard of 1 μgZENeq / kg. In Example 2, due to the removal of micro-etching, the extractable ZEN in step three increased to 14 μg / kg, the total degradation rate decreased to 91.5%, and the EEC increased to 1.2 μgZENeq / kg. In Example 4, due to a fixed ventilation rate, free ZEN decreased to only 48 μg / kg in step two, and extractable ZEN increased to 18 μg / kg in step three. The total degradation rate decreased to 88.3%, and the EEC reached 2.1 μgZENeq / kg. Example 6 passed chemical testing but exceeded the EEC standard (see Table 4 for post-ripening treatment details). In Comparative Example 1, step two yielded a high level of free ZEN (85 μg / kg), with a total degradation rate of only 68.5%, a conjugated ZEN residue of 8.2 μg / kg, and an EEC of 4.8 μgZENeq / kg. In Comparative Example 2, due to the drying of raw materials and the implementation of a detoxification strategy, the conjugated ZEN residue was 28.5 μg / kg, with a total degradation rate of only 52.0% and an EEC of 5.2 μgZENeq / kg. Comparative Example 3, by directly omitting the enzymatic hydrolysis step, resulted in a conjugated ZEN residue as high as 35.8 μg / kg, the highest value in the entire table, with a total degradation rate of only 78.6% and an EEC of 3.1 μgZENeq / kg. Although Comparative Example 4 met the chemical indicators (α-ZOL + ZAN 3.0 μg / kg ≤ 5), its EEC reached 1.1 μgZENeq / kg, posing a risk of missed detection. This indicates that the enzymatic release of bound ZEN in step one is a necessary preliminary step in the entire detoxification chain. The synergistic effect of micro-etching, dynamic ventilation, and targeted biodegradation has a significant impact on improving the total ZEN degradation rate, and relying solely on chemical detection cannot guarantee the biosafety of breeding cows.

[0075] Table 4 Comparison of the effects of post-ripening treatment when EEC levels exceed the standard

[0076] As shown in Table 4, only Example 6 triggered in-situ ripening treatment after EEC exceeded the standard. This batch passed chemical testing (α-ZOL + ZAN 4.5 μg / kg, free ZEN not detected), but the initial EEC value was 1.3 μg ZENeq / kg (>1 μg ZENeq / kg). After adding 0.15% stevia polyphenols (using a 10% polyphenol solution atomized spray, resulting in a weight gain of 1.5%) and ripening for 2.5 days under sealed conditions with ORP=-220mV, the EEC decreased to 0.4 μg ZENeq / kg, and the stevia polyphenol residue rate was 65% (≥60%), achieving the target biopotency. The other examples did not trigger ripening because both chemical and EEC tests were satisfactory; data for the comparative examples are not available in this table because no EEC quality control or ripening mechanism was implemented. This demonstrates that the in-situ post-ripening mechanism can achieve the target biopotency by targeted delivery of polyphenols and a strong reducing environment without returning the product to the container, providing a feasible quality control rework path for batch isolation management in continuous production.

[0077] Table 5 Comparison of the composition, nutritional quality, and physiological compatibility of anaerobic fermentation products with cows

[0078] As shown in Table 5, the anaerobic fermentation product composition of Example 1 was optimal: the pH decrease rate from 0 to 48 hours was 0.275 / day, the pH fluctuation range after 48 hours was ≤0.15, the volatile fatty acid molar composition was acetic acid:propionic acid:butyric acid = 7.0:2.4:0.6, the γ-aminobutyric acid content reached 38 mg / kg, the crude protein increase rate was 28%, the acid-soluble protein was 16.5%, and the fiber degradation rate was 52%. In the cow feeding verification, the daily pH fluctuation range of the rumen was ±0.12, which was in a very stable state. In Example 3, because no staged pH control was performed, the pH decrease rate from 0 to 48 hours reached 0.65 / day, the fluctuation range in the later stage was ±0.45, the butyric acid ratio increased to 9%, and the GABA content decreased to 22 mg / kg, which was lower than the lower limit requirement of 30 mg / kg. In Example 5, because stevia polyphenols and calcium bicarbonate were removed, the pH fluctuation range reached ±0.25, the GABA content decreased to 28 mg / kg, and the crude protein increase rate and fiber degradation rate both decreased significantly. Example 12, under the lower limit of phased parameters (pH decrease rate of 0.3 / day from 0 to 48 h), with a VFA molar composition of 6.0:2.0:0.5 and a GABA content of 32 mg / kg, still met the rumen adaptability requirements of cows. Comparative Examples 1 and 2, due to traditional processes and dried raw materials, showed crude protein increases of only 18% and 20%, respectively, and significantly lower rates of acid-soluble protein and fiber degradation compared to Example 1. Comparative Example 4, using traditional strong acid fermentation, had a VFA composition of 5.0:2.5:0.8, a butyric acid content of 10%, and a GABA content of only 30 mg / kg. The daily rumen pH fluctuation range reached ±0.35, significantly worse than Example 1. This indicates that phased weak acid fermentation and the synergistic addition of stevia polyphenols and calcium bicarbonate have significant effects on targeted regulation of VFA composition, increasing GABA content, improving nutritional quality, and stabilizing the rumen pH environment of cows. Traditional strong acid fermentation schemes cannot be directly applied to pregnant cows in late pregnancy and peripartum periods with zero tolerance for juvenile ketones.

[0079] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for preparing stevia residue feed, characterized in that, Includes the following steps: Step 1: The fresh wet stevia residue after stevia glycoside extraction is subjected to differential particle size pulverization to form coarse, medium and fine powder segments. The coarse, medium and fine powder segments are pretreated with differential moisture content. The fine powder segment is then mixed with fresh wet corn steep liquor containing bound and free gibberellic acid. The bound gibberellic acid is released by enzymatic hydrolysis. The mixture is then mixed with the coarse and medium powder segments and tomato cake meal as an oily protein adjuvant. The mass ratio of stevia residue, tomato cake meal and corn steep liquor is controlled at (75~80):(15~20):(3~5). The oil content of the mixture is controlled at 4~8%. A water-in-oil crude emulsion system is constructed to enrich the fat-soluble gibberellic acid at the oil droplet interface, thus obtaining the pretreated material. Step 2: Inoculate the pretreated material obtained in Step 1 with enzyme-producing microorganisms containing Aspergillus niger, and carry out aerobic pre-detoxification treatment under aeration conditions to perform targeted biodegradation of the existing and enzymatically released juvenile ketones, thereby degrading the free juvenile ketones and obtaining the detoxified material. Step 3: Inoculate the detoxified material obtained in Step 2 with compound microbial agent and carry out anaerobic fermentation treatment. Control the pH of the fermentation endpoint to a weakly acidic environment of 4.8~5.2 to obtain fermented roughage. Step 4: Detect the chemical content of zearalenone, α-zearalenone alcohol, zearalenone and their conjugated compounds in fermented roughage, and use biological effect tests to detect their estrogen activity equivalent. The product is released only if both the chemical content and the estrogen activity equivalent meet the safety standards. The stevia residue feed is suitable for pregnant cows in late pregnancy and peripartum period who have zero tolerance for juvenile ketone.

2. The method for preparing a stevia residue feed according to claim 1, characterized in that, In step one, the coarse particle size is 10-15mm, the medium particle size is 6-8mm, and the fine particle size is ≤2mm. The mass ratio of the three is (20-30):(50-60):(10-20), and the sum of the mass parts of the three is 100 parts. The moisture content of the coarse powder segment is 60-70%, the moisture content of the medium powder segment is 60-65%, and the moisture content of the fine powder segment is 55-60%. The moisture content of fresh wet stevia residue is 60-70%, and the moisture content of fresh wet corn liquor is 45-55%.

3. The method for preparing a stevia residue feed according to claim 1, characterized in that, In step one, the stevia residue includes coarse segments with a particle size of 10-15 mm. After differential moisture content pretreatment, the moisture content of the coarse segments is 60-68%. Then, it is pretreated by atomized spraying with a calcium bicarbonate saturated solution, controlling the weight gain of the coarse segments to be 3-5%. It is then left to stand at 25-30℃ for 2-4 hours to form a microporous etching structure with a depth of 50-200 μm on the surface of the coarse segments, and the moisture content of the coarse segments after spraying is controlled to not exceed 70%. The oil content of tomato cake meal is 25-30%, which forms an oil-in-water crude emulsion system with the fibrous phase of stevia residue and the aqueous phase of corn syrup.

4. The method for preparing a stevia residue feed according to claim 1, characterized in that, In step two, the enzyme-producing microorganisms are a compound inoculum of Aspergillus niger and Bacillus subtilis, wherein the viable Aspergillus niger count is 1×10⁻⁶. 8 ~5×10 8 CFU / g dry weight of material, the number of viable Bacillus subtilis is 20 to 60 times that of viable Aspergillus niger; the aerobic pre-detoxification treatment opens the benzodifuran ring of jugazocarpine through extracellular laccase and peroxidase secreted by Aspergillus niger, and breaks its lactone bond through lactone hydrolase of Bacillus subtilis, so that jugazocarpine is degraded into metabolites without estrogen activity. The degradation targets are the free gibberellenone already present in the material and the bound gibberellenone released by enzymatic decomposition in step one. The conditions for aerobic pre-detoxification treatment are: ventilation rate 0.5~1.0 vvm, temperature 32~35℃, and reaction time 2.0~2.5h; the viable bacteria count is based on the dry weight of the pretreated material obtained in step one.

5. The method for preparing a stevia residue feed according to claim 1, characterized in that, The aeration rate mentioned in step two is dynamically positively correlated with the amount of corn steep liquor added in step one. When the mass percentage of corn steep liquor is 3-5%, the following condition is met: The ventilation rate is calculated as 0.5 + 0.1 × (corn steep liquor mass percentage - 3%), and the ventilation rate does not exceed 1.0 vvm.

6. The method for preparing a stevia residue feed according to claim 1, characterized in that, The compound microbial agent in step three consists of Lactobacillus plantarum and Saccharomyces cerevisiae, with Lactobacillus plantarum having a viable count of 7 × 10⁻⁶. 9 ~9×10 9 The CFU / g dry weight of the material and the viable count of brewer's yeast were 3×10¹¹~4.5×10¹¹ CFU / g dry weight of the material, with a viable count ratio of 1:(30~70); the anaerobic fermentation treatment was carried out at 28~32℃ for 5~7 days; in step three, 0.2~0.5% stevia polyphenols and 1.5~2.5% calcium bicarbonate were also added; the viable counts were all based on the dry weight of the pretreated material obtained in step one.

7. The method for preparing a stevia residue feed according to claim 1, characterized in that, At the end of step two, the content of free gibberellic acid was detected and controlled to be ≤50μg / kg; At the end of step three, the content of extractable jugaprone was detected and controlled to be ≤10μg / kg; In step four, when the product is released, the sum of the contents of α-zearalenone and zearalenone is tested and controlled to be ≤5μg / kg, and free zearalenone is not detected.

8. The method for preparing a stevia residue feed according to claim 1, characterized in that, Step four involves using a biological effects assay to determine the estrogen activity equivalent, specifically: The estrogenic activity equivalent of the fermented roughage extract was determined by a recombinant yeast estrogen screening test, and the estrogenic activity equivalent ≤1μgZENeq / kg was used as an additional release standard. The determination of estrogen activity equivalent includes a pre-hydrolysis step, in which the extract is pretreated with β-glucuronidase and sulfatase to release the conjugated estrogen active substances before the total estrogen activity equivalent is determined. The conjugated states include gibbenone-14-β-D-glucoside and gibbenone sulfate; If the chemical test is qualified but the estrogen activity equivalent exceeds the standard, 0.1-0.2% stevia polyphenols should be added to the batch of fermented roughage, and anaerobic ripening treatment should be continued for 2-3 days under sealed conditions until the estrogen activity equivalent reaches the standard.

9. The method for preparing a stevia residue feed according to claim 1, characterized in that, The initial pH value is 5.8-6.5, which is the pH of the material after the aerobic pre-detoxification in step two. The anaerobic fermentation treatment in step three is divided into two stages: The first stage is a rapid acid reduction period of 0~48h, during which the pH is controlled to decrease from the initial value to 5.0~5.2, and the pH decrease rate is 0.3~0.5 / day; The second stage is the stable fermentation period after 48 hours, during which the pH fluctuation range is controlled to be ≤0.2, and the pH is maintained at 4.8~5.0; The final fermentation pH of 4.8-5.2 is higher than that of conventional silage (pH 4.0-4.5). The weak acid preservation is achieved through the synergistic antibacterial effect of stevia polyphenols and lycopene. The molar composition of volatile fatty acids in fermented roughage is acetic acid: propionic acid: butyric acid = (6-8): (2-3): (0.5-1), and the content of γ-aminobutyric acid is ≥30mg / kg.

10. The method for preparing a stevia residue feed according to claim 8, characterized in that, The method for supplementing stevia polyphenols is as follows: Stevia polyphenols were dissolved in a weakly alkaline buffer solution with a pH of 7.0-7.5 to prepare a 10% polyphenol solution. The solution was then evenly applied to the surface of fermented roughage using a mist spraying method, with the weight gain rate controlled at 1-2%. Anaerobic post-ripening treatment under closed conditions controls the oxidation-reduction potential (ORP) to be ≤-200mV, and the residual rate of stevia polyphenols in fermented roughage after post-ripening is ≥60%.

Images